WO2014169884A2 - 动态带宽调度方法、装置及计算机存储介质 - Google Patents
动态带宽调度方法、装置及计算机存储介质 Download PDFInfo
- Publication number
- WO2014169884A2 WO2014169884A2 PCT/CN2014/080027 CN2014080027W WO2014169884A2 WO 2014169884 A2 WO2014169884 A2 WO 2014169884A2 CN 2014080027 W CN2014080027 W CN 2014080027W WO 2014169884 A2 WO2014169884 A2 WO 2014169884A2
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- bandwidth
- column
- row
- node
- matrix
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 90
- 238000003860 storage Methods 0.000 title claims abstract description 18
- 230000003287 optical effect Effects 0.000 claims abstract description 38
- 239000011159 matrix material Substances 0.000 claims description 262
- 238000013475 authorization Methods 0.000 claims description 203
- 230000009467 reduction Effects 0.000 claims description 100
- 230000005540 biological transmission Effects 0.000 claims description 48
- 235000008694 Humulus lupulus Nutrition 0.000 claims description 21
- 230000015572 biosynthetic process Effects 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 12
- 238000003786 synthesis reaction Methods 0.000 claims description 12
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 230000001934 delay Effects 0.000 claims description 3
- 238000009966 trimming Methods 0.000 claims 13
- 238000013138 pruning Methods 0.000 claims 11
- 239000013307 optical fiber Substances 0.000 claims 1
- 230000002194 synthesizing effect Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 7
- 239000000835 fiber Substances 0.000 description 6
- 230000007246 mechanism Effects 0.000 description 6
- 238000012913 prioritisation Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000005520 cutting process Methods 0.000 description 4
- 230000003111 delayed effect Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000011946 reduction process Methods 0.000 description 2
- 229920006395 saturated elastomer Polymers 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 235000019580 granularity Nutrition 0.000 description 1
- 235000003642 hunger Nutrition 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000037351 starvation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/28—Flow control; Congestion control in relation to timing considerations
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04Q—SELECTING
- H04Q11/00—Selecting arrangements for multiplex systems
- H04Q11/0001—Selecting arrangements for multiplex systems using optical switching
- H04Q11/0062—Network aspects
- H04Q11/0066—Provisions for optical burst or packet networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/12—Discovery or management of network topologies
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/20—Hop count for routing purposes, e.g. TTL
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04Q—SELECTING
- H04Q11/00—Selecting arrangements for multiplex systems
- H04Q11/0001—Selecting arrangements for multiplex systems using optical switching
- H04Q11/0062—Network aspects
- H04Q2011/0064—Arbitration, scheduling or medium access control aspects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04Q—SELECTING
- H04Q11/00—Selecting arrangements for multiplex systems
- H04Q11/0001—Selecting arrangements for multiplex systems using optical switching
- H04Q11/0062—Network aspects
- H04Q2011/009—Topology aspects
- H04Q2011/0092—Ring
Definitions
- the present invention relates to the field of network communication technologies, and in particular, to a dynamic bandwidth scheduling method, apparatus, and computer storage medium. Background technique
- IP Internet Protocol
- the new Optical Burst Transport Ring-Network can not only be friendly to the burstiness of data services, but also effectively reduce the demanding requirements of all-optical networks for optical devices. It has flexible networking capabilities. And full feasibility.
- the network topology of OBTN is shown in Figure 1, where the data channel consists of several wavelengths (where the wavelengths are: ⁇ 0, ⁇ 1, -.. ⁇ ⁇ , where ⁇ is a positive integer), which is used to carry optical burst data.
- the control channel uses independent wavelength ⁇ c for carrying control information such as bandwidth allocation and time slot synchronization; the data frame and the control frame are separately processed at each network node, and the network node performs corresponding service reception on the data frame according to the control information, Send operation.
- each network node of the OBTN employs a fixed wavelength transmitter, a tunable wavelength receiver.
- OBTN uses burst as the smallest network switching unit.
- Each burst consists of a control packet (BCP, Burst Control Packet) and a burst data packet (BDP, Burst Data Packet), both of which are on the physical channel.
- BCP Burst Control Packet
- BDP Burst Data Packet
- optical burst transmission ring networks generally use token access or random access based on collision detection. From a technical perspective, both are distributed control management. However, distributed MAC technology can easily cause conflicts of burst data packets, resulting in wasted bandwidth resources of the optical burst ring network. In addition, since the distributed medium access technology additionally adds an appropriate control protocol to provide fairness guarantee, the processing delay of the control information by the network node is preferably increased.
- MAC media access control
- centralized MAC technology based on slot division can improve network performance better: Select one node as the master node in the ring network, and the other nodes as slave nodes.
- the master node is responsible for controlling the main functions of the plane, including: collecting bandwidth requests of the entire network node, allocating bandwidth according to related resources, updating and delivering bandwidth maps, etc.; the slave node performs corresponding receiving and transmitting operations according to the bandwidth map delivered by the master node.
- all bandwidth scheduling policies are controlled by the master node, which can globally coordinate network resources, maximize bandwidth utilization, provide fairness guarantees, and reduce conflicting contention.
- the cross-master service refers to a data packet that needs to pass through the master node before being sent from the source node to the corresponding destination node; and the non-cross-node service refers to before the source node is sent to the corresponding destination node to go down the road. There is no need to go through the data packets of the primary node.
- the slave node Node5 sends the data service to the destination node Node3 on the time slot m according to the current bandwidth map.
- the control frame including the bandwidth map re-arrives to the master node Node1.
- the primary node performs bandwidth allocation based on the new bandwidth request.
- Node2 will probably be based on the most
- the new bandwidth map is also sent to the data service of the destination node Node3 on the time slot m.
- the data service sent by the Node 5 to the Node 3 on the data frame slot m is still not in the road, and the two are bound to cause a collision.
- Node2 does not send data to Node3 on the time slot m, when the control frame and the data frame arrive at Node3, Node3 will receive the service according to the new bandwidth map, and it is also possible that Node5 is not available in time slot T1.
- the embodiment of the invention provides a dynamic bandwidth scheduling method, device and computer storage medium, which can be applied to an optical burst transmission ring network, which can solve the dynamic resource scheduling conflict of the optical burst transmission ring network based on time slot division and centralized control. problem.
- An embodiment of the present invention provides a dynamic bandwidth scheduling method, which is applied to a node in an OBTN, and includes:
- the target node When the target node is the master node, for each source node in the OBTN, when the target node allocates a time slot for the connection from the source node to the destination node, when there is a burst time slot OB that has not been dropped In the slot, the time slot occupied by the destination node with the smallest number of hops of the currently configured destination node is preferentially selected;
- the slot allocation result is converted into a bandwidth map and sent to each slave node in the OBTN.
- bandwidth authorization is performed for the connection between each source node and the destination node according to the bandwidth request of the entire network node.
- the bandwidth authorization is performed for the connection between each source node and the destination node according to the bandwidth request of the node of the entire network, including:
- the master node obtains a bandwidth request matrix according to the bandwidth request of the entire network node, where an element in the bandwidth request matrix indicates a bandwidth request of the node i to the node j in the OBTN, where i and j are respectively less than or equal to N.
- N is the total number of nodes in the OBTN;
- the master node performs the reduction authorization on the bandwidth request matrix according to the existing bandwidth resource to obtain a bandwidth authorization matrix.
- the element in the bandwidth authorization matrix represents the authorized bandwidth of the node i to the node j in the OBTN, and i and j are positive integers less than or equal to N, respectively, where N is the total number of nodes in the OBTN;
- the sum of the authorized bandwidths of each row element is less than or equal to the maximum line rate; for each column in the bandwidth grant matrix, the sum of the authorized bandwidths of all elements in the upper triangular portion of the column, and before a loop length period The sum of authorized bandwidths of all elements in the lower triangular portion of the same column in the received bandwidth grant matrix, less than or equal to the maximum line rate.
- the primary node performs the reduction authorization on the bandwidth request matrix according to the existing bandwidth resource, including:
- the master node performs row clipping on the bandwidth request matrix; the master node performs column reduction on the matrix obtained after the row is reduced.
- the performing, by the primary node, performing row clipping on the bandwidth request matrix includes: determining, for each row in the bandwidth request matrix, whether a sum of elements in the row exceeds the maximum line rate; if Re-allocating the bandwidth of each element in the row until the sum of the elements in the row is less than or equal to the maximum line rate;
- the primary node performs column reduction on the matrix obtained after the row is cut, and includes: determining, for each row in the matrix obtained by the row reduction, determining the sum of each element in the lower triangular portion of the column Whether the difference between the maximum line rate and the reserved bandwidth reserved for the column is exceeded; if it is exceeded, the bandwidth allocation is performed on each element in the lower triangular portion of the column until the resulting lower triangle of the column The sum of the elements in the portion is less than or equal to the difference between the maximum line rate and the reserved bandwidth reserved for the column; on the other hand, determining whether the sum of the elements in the upper triangular portion of the column exceeds The maximum bandwidth grant amount of the upper triangular portion of the column; if it is exceeded, the bandwidth allocation is performed on each element in the upper triangular portion of the column until the obtained column is The sum of the elements in the upper triangular portion is less than or equal to the maximum bandwidth grant amount of the upper triangular portion of the column; wherein, the maximum bandwidth grant amount of the upper triangular
- the master node performs column clipping on the bandwidth request matrix, including: determining, for each column in the bandwidth request matrix, determining whether a sum of each element in a lower triangular portion of the column exceeds a maximum The difference between the line rate and the reserved bandwidth reserved for the column; if exceeded, the bandwidth allocation is performed on each element in the column until the sum of the elements in the lower triangular portion of the column is obtained, Less than or equal to the difference between the maximum line rate and the reserved bandwidth reserved for the column; on the other hand, determining whether the sum of each element in the upper triangular portion of the column exceeds the upper triangular portion of the column The maximum bandwidth grant amount; if exceeded, the bandwidth allocation is performed on each element in the upper triangular portion of the column until the sum of the elements in the upper triangular portion of the obtained column is less than or equal to the upper triangle of the column a maximum bandwidth grant amount of the portion; wherein a maximum bandwidth grant amount of the upper triangular portion of the column is the maximum line rate and
- Determining, by the master node, the matrix obtained after the column reduction includes: determining, for each row in the matrix obtained by the column reduction, whether the sum of each element in the row exceeds the maximum line rate; If it is exceeded, the bandwidth allocation is performed on each element in the row until the sum of the elements in the obtained row is less than or equal to the maximum line rate.
- the performing, by the primary node, performing row clipping on the bandwidth request matrix includes: determining, for each row in the bandwidth request matrix, whether a sum of elements in the row exceeds the maximum line rate; if Re-allocating the bandwidth of each element in the row until the sum of the elements in the row is less than or equal to the maximum line rate;
- the master node performs column reduction on the bandwidth request matrix, including:
- the method further includes:
- the primary node After determining the value of each element in the bandwidth authorization matrix, the primary node scans the bandwidth authorization matrix row by row and column by column, for a row whose sum of elements in the row is less than the maximum line rate And increasing the value of the element at the intersection of the sum of the elements in the upper triangular portion of the column and the column of the maximum bandwidth grant amount smaller than the upper triangular portion of the column, and after the addition, each of the elements in the row The sum of the elements does not exceed the maximum line rate, the sum of the elements in the upper triangular portion of the column in which the element is located does not exceed the maximum bandwidth grant amount of the upper triangular portion of the column; for the intersection of the row and the column The value of the element is increased, the behavior includes a row of each element that is smaller than the maximum line rate, and the column satisfies: a sum of elements in a lower triangular portion of the bandwidth authorization matrix, less than the maximum The difference between the line rate and the reserved bandwidth reserved for the column. After the addition, the
- the re-allocating the bandwidth of each element in the column includes: the highest priority according to the fixed bandwidth allocation mode, the second priority of the guaranteed bandwidth allocation mode, and the lowest priority of the non-guaranteed bandwidth allocation mode.
- the order of bandwidth allocation for each element in the column correspondingly with the highest priority bandwidth allocation currently available;
- Re-allocating the bandwidth of each element in the row including:
- the currently available highest priority bandwidth allocation manner is correspondingly Each element in the row is allocated bandwidth.
- the method further includes:
- bandwidth allocation is performed on the elements in the column or the row by using a non-guaranteed bandwidth allocation manner, if there is still remaining bandwidth in the column or row, the remaining bandwidth polling is correspondingly allocated to the column or row.
- one OB is allocated per poll; where the unsaturation is a connection with an existing bandwidth grant less than the actual bandwidth request.
- the method further includes:
- the master node When the time slot allocation is performed, the master node performs time slot configuration in sequence from the master node in sequence according to the node arrangement order in the OBTN ring;
- the order of configuration of each destination node in each of the source nodes is as follows: According to the order of nodes in the OBTN ring, time slot configuration is performed in order of the number of hops of each destination node from the source node in descending order.
- the method further includes:
- the time slot with the smallest slot number is preferred.
- the method further includes:
- the target node When the target node is a slave node, the target node synthesizes the received bandwidth map and the cross-ring bandwidth map and obtains a received map, and receives the corresponding data frame according to the received map;
- the cross-ring bandwidth map is a bandwidth map received by the target node before a loop length period; in the receiving map, the element RnnH indicates that the destination node 11 receives the OB sent by the source node j in the slot m; In the map, the element AnnH indicates that the source node n sends the OB to the destination node i in the time slot m; m, n, i, and j are all positive integers; wherein, the bandwidth map and the cross-ring bandwidth map received by the current time are Synthesize to get the received map, including:
- the method further includes:
- the node delays the received data frame by using a fiber delay line; wherein, the delay time is greater than or equal to the control node of the local node Processing time.
- the embodiment of the present invention further provides a dynamic bandwidth scheduling apparatus, which is applied to a main node in an OBTN, and includes:
- the time slot configuration module is configured to allocate, for each source node in the OBTN, a time slot for the connection from the source node to the destination node, and in each time slot where there is a burst time slot OB that has not been dropped, the priority is selected.
- the bandwidth map conversion module is configured to convert the time slot allocation result obtained by the time slot configuration module into a bandwidth map and send it to each slave node in the OBTN.
- the method further comprises:
- a bandwidth authorization matrix generating module configured to obtain a bandwidth request matrix according to a bandwidth request of the entire network node, where an element in the bandwidth request matrix indicates a bandwidth request of the node i to the node j in the OBTN, where i and j are respectively a positive integer less than or equal to N, where N is the total number of nodes in the OBTN;
- the reduction authorization module is configured to perform the reduction authorization on the bandwidth request matrix according to the existing bandwidth resource to obtain a bandwidth authorization matrix, where an element in the bandwidth authorization matrix indicates an authorized bandwidth of the node i to the node j in the OBTN, I and j are positive integers less than or equal to N, respectively, N is the total number of nodes in the OBTN; in the bandwidth authorization matrix, the sum of the authorized bandwidths of each row element is less than or equal to the maximum line rate; For each column, the sum of the authorized bandwidths of all elements in the upper triangular portion of the column and the sum of the authorized bandwidths of all elements in the lower triangular portion of the same column in the bandwidth grant matrix received before a ring length period, less than or equal to The maximum line rate is described.
- the cut authorization module includes:
- a row reduction unit configured to perform line reduction on the bandwidth request matrix
- Column reduction unit configured to perform column reduction on the matrix obtained after the row is cut.
- the line cut unit includes: a first determining subunit, configured to determine, for each row in the bandwidth request matrix, whether a sum of elements in the row exceeds the maximum line rate; a unit configured to perform bandwidth allocation on each element in the row when the determination result of the first determining subunit is exceeded, until the sum of the elements in the obtained row is less than or equal to the maximum line Rate
- the column clipping unit includes: a second determining subunit configured to determine, for each column in the matrix obtained after the row clipping, whether the sum of each element in the lower triangular portion of the column is A difference between a maximum line rate and a reserved bandwidth reserved for the column; a second allocation subunit configured to re-down the lower triangle of the column when the judgment result of the second determining subunit is exceeded Each element in the portion performs bandwidth allocation until the sum of the elements in the lower triangular portion of the obtained column is less than or equal to the difference between the maximum line rate and the reserved bandwidth reserved for the column;
- a third determining subunit configured to determine, for each column in the matrix obtained after the row is cut, whether a sum of elements in the upper triangular portion of the column exceeds a maximum bandwidth authorization of an upper triangular portion of the column
- a third allocation subunit configured to, when the judgment result of the third determining subunit is exceeded, perform bandwidth allocation on each element in the upper triangular part of the column until the obtained column is The sum of the elements in the upper triangular portion is less than or equal to the maximum bandwidth grant amount of the upper triangular portion of the column;
- the maximum bandwidth grant amount of the upper triangular portion of the column is the difference between the maximum line rate and the actual bandwidth grant amount of the lower triangle portion of the same column in the bandwidth grant matrix received before one loop length period.
- the line reduction unit includes: a fourth determining subunit, configured to determine, for each row in the matrix obtained after the column reduction, whether a sum of each element in the row exceeds the maximum line rate; a fourth allocation subunit, configured to perform bandwidth allocation on each element in the row when the judgment result of the fourth judging subunit is exceeded, until the sum of the elements in the obtained row is less than or equal to The maximum line rate; wherein
- the column clipping unit includes: a fifth determining subunit configured to determine, for each column in the bandwidth request matrix, whether a sum of elements in a lower triangular portion of the column exceeds a maximum line rate and is the column The difference between the preset bandwidth reservations; and the fifth allocation subunit, configured to perform bandwidth allocation on each element in the column when the judgment result of the third determining subunit is exceeded, until the obtained location
- the sum of the elements in the lower triangular portion of the list is less than or equal to the difference between the maximum line rate and the reserved bandwidth reserved for the column;
- a sixth determining subunit configured to determine, for each column in the bandwidth request matrix Whether the sum of each element in the upper triangular portion of the list exceeds the maximum bandwidth grant amount of the upper triangular portion of the column; the sixth allocation subunit is configured to, when the judgment result of the fifth determining subunit is exceeded, Re-allocating the elements in the upper triangular portion of the column until the sum of the elements in the upper triangular portion of the column is less than or equal to the maximum bandwidth of the upper triangular portion of the column;
- the maximum bandwidth grant amount of the upper triangular portion of the list is the difference between the maximum line rate and the actual bandwidth grant amount of the lower triangular portion of the same column in the bandwidth grant matrix received before one loop long period.
- the cut authorization module further includes:
- the authorization synthesizing unit is configured to: when determining the bandwidth authorization matrix, take the smaller one of the values of the matrix obtained after the row reduction and the same position in the matrix obtained after the column reduction, as the bandwidth authorization matrix The value of the element in the corresponding position;
- the row clipping unit includes: a seventh determining subunit configured to determine, for each row in the bandwidth request matrix, whether a sum of elements in the row exceeds the maximum line rate; a seventh allocation subunit, configured When the judgment result of the seventh judging subunit is exceeded, bandwidth allocation is performed on each element in the row until the sum of the elements in the obtained row is less than or equal to the maximum line rate;
- the column clipping unit includes: an eighth determining subunit configured to determine, for each column in the bandwidth request matrix, whether a sum of elements in a lower triangular portion of the column exceeds a maximum line rate and is the column a difference between the preset bandwidth reservations; and an eighth allocation subunit configured to perform bandwidth allocation on each element in the lower triangular portion of the column when the judgment result of the eighth determining subunit is exceeded, Up to the obtained sum of elements in the lower triangular portion of the column being less than or equal to a difference between the maximum line rate and a reserved bandwidth reserved for the column;
- a ninth determining subunit configured to determine, for each column in the bandwidth request matrix, whether a sum of elements in an upper triangular portion of the column exceeds a maximum bandwidth grant amount of an upper triangular portion of the column; Assigning a subunit, configured to determine that the ninth determining subunit is When exceeded, bandwidth allocation is performed on each element in the upper triangular portion of the column until the sum of the elements in the upper triangular portion of the obtained column is less than or equal to the maximum bandwidth authorization amount of the upper triangular portion of the column;
- the maximum bandwidth grant amount of the upper triangular portion of the column is the difference between the maximum line rate and the actual bandwidth grant amount of the lower triangle portion of the same column in the bandwidth grant matrix received before one loop long period.
- the authorization synthesizing unit comprises:
- the scanning subunit is configured to scan the bandwidth authorization matrix row by row and column by column after determining the value of each element in the bandwidth authorization matrix;
- a first increasing sub-unit configured to be a column having a maximum bandwidth grant amount smaller than a sum of each element in the upper triangular portion of the row and the column in the row and the column in the row is smaller than the upper triangular portion of the column for the sum of the elements in the row
- the value of the element at the intersection is increased.
- a second increasing subunit configured to have a sum of elements of the row that are smaller than the maximum line rate and a sum of elements in the lower triangular portion of the column that are smaller than the maximum line rate and a bandwidth preset for the column.
- the time slot configuration module is further configured to: when the time slot is allocated, according to the node arrangement order in the OBTN ring, sequentially perform the time slot configuration from the node as the source node in sequence; and further configured to determine The configuration order of each destination node in each source node is: according to the order of nodes in the OBTN ring, the slot configuration sequence is sequentially performed according to the hop count of each destination node from the source node.
- the time slot configuration module is further configured to be a source node to a destination node
- the time slot is preferred. The smallest time slot.
- the embodiment of the invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the dynamic bandwidth scheduling method described above.
- an efficient and conflict-free dynamic resource scheduling of the optical burst transmission ring network can be realized, and the conflict problem of the cross-master service can be completely solved without interrupting the service.
- the invention can not only allocate bandwidth resources reasonably and reasonably, but also quickly respond to the bandwidth requirement of the burst service, and can realize data collision-free switching, channel space reuse, strict quality of service (QoS) guarantee, and high bandwidth utilization. rate. DRAWINGS
- FIG. 1 is a schematic diagram of an optical burst transmission ring network in the related art
- FIG. 2 is a topological structural diagram of a 5-node optical burst transmission ring network in the related art
- FIG. 3 is a flowchart of a dynamic bandwidth scheduling method of an optical burst transmission ring network according to an embodiment of the present invention
- FIG. 4 is a graphical representation of a data frame and a bandwidth map according to an embodiment of the present invention
- FIG. 5 is a flowchart of a method for scheduling dynamic bandwidth of an optical burst ring network
- FIG. 6 is a schematic diagram of a process of synthesizing a received bandwidth map according to an embodiment of the present invention.
- FIG. 7 is a structural diagram of a dynamic bandwidth scheduling apparatus of an optical burst transmission ring network according to an embodiment of the present invention
- FIG. 8 is a schematic diagram of a cut authorization module according to an embodiment of the present invention
- FIG. 9a and 9b are flowcharts of a line-cutting process and a column-cutting process in an embodiment of the present invention
- FIG. 10 is a schematic diagram showing an area of a service between a primary node and a non-cross-node in a bandwidth authorization matrix according to an embodiment of the present invention
- FIG. 11 is a flow chart of authorization synthesis in an embodiment of the present invention.
- FIG. 12 is a schematic diagram of a time slot configuration module in an embodiment of the present invention
- 13a and 13b are respectively an example of time slot prioritization and a time slot selection priority graphical representation according to an embodiment of the present invention
- 15a to 15d are structural diagrams of a dynamic bandwidth scheduling apparatus in an embodiment of the present invention. detailed description
- the embodiment of the invention describes a dynamic bandwidth scheduling method for an optical burst transmission ring network, which is applied to each node in the OBTN. As shown in FIG. 3, the method includes:
- Step 301 When the target node is used as the primary node, for each source node in the OBTN, when a time slot is allocated for the connection from the source node to a destination node, there is a burst slot that has not been dropped ( In each time slot of the OB, Optical Burst, the time slot occupied by the destination node with the smallest number of hops of the currently configured destination node is preferentially selected;
- Step 302 When the target node is the master node, the time slot allocation result is converted into a bandwidth map and sent to each slave node in the OBTN.
- bandwidth authorization is performed for the connection between each source node and the destination node according to the bandwidth request of the entire network node.
- the bandwidth authorization for the connection between each source node and the destination node according to the bandwidth request of the entire network node may be implemented by the following steps:
- Step 10 The master node (that is, the target node) combines the bandwidth requests of the slave nodes obtained from the control frame information to obtain a bandwidth request matrix.
- the element in the bandwidth request matrix represents a bandwidth request of the node i to the node j in the OBTN, where i and j are positive integers equal to or less than N, respectively, and N is the total number of nodes in the OBTN.
- the i-th row element represents the node i in the OBTN
- the i-th column element represents a bandwidth request of each node in the OBTN to the node 1;
- Step 20 The primary node performs the reduction authorization on the bandwidth request matrix according to the existing bandwidth resource to obtain a bandwidth authorization matrix.
- the bandwidth authorization matrix determines the size of the authorized bandwidth between any two nodes in the OBTN, that is, the number of authorized OBs.
- the reduction authorization includes row reduction and column reduction, and the row reduction is: performing bandwidth constraint on the bandwidth request matrix row by row. When the total bandwidth request of a row exceeds the maximum line rate of the row, the elements in the row are required.
- the bandwidth request matrix may be divided into an upper triangular portion and a lower triangular portion; wherein, the upper triangular portion is a non-cross-primary node
- the bandwidth request of the service, the lower triangle part is the bandwidth request across the service of the primary node, and accordingly, the column is reduced to: For each column in the bandwidth request matrix, the total bandwidth request and the lower triangle of the upper triangular part in the column
- the total bandwidth request is partially bandwidth-constrained.
- the available bandwidth allocation methods include: fixed bandwidth allocation, guaranteed bandwidth allocation, and non-guaranteed bandwidth allocation to provide sufficient fairness guarantee and QoS guarantee.
- the maximum bandwidth grant amount of the lower triangle portion in the column is the difference between the maximum line rate and the reserved bandwidth reserved for the column; the upper triangular portion of the column
- the maximum bandwidth grant amount is the difference between the maximum line rate and the actual bandwidth grant amount of the lower triangle portion of the same column in the bandwidth grant matrix received before one loop long period;
- Step 30 The primary node maps the bandwidth authorization matrix to a time slot by using a time slot configuration method.
- the allocation table the bandwidth map.
- the form of the bandwidth map is a two-dimensional matrix. As shown in FIG. 4, assuming that the value of a certain element B in the two-dimensional matrix is k, it indicates that the node m has an OB addressed to the node k on the slot n.
- the time slot configuration method is the "sequential configuration" method.
- the core idea is: (1) Configuring the source nodes in sequence, according to the order of nodes in the OBTN, sequentially setting each node as a source node in sequence from the primary node.
- the embodiment of the invention further describes a dynamic bandwidth scheduling device for an optical burst transmission ring network, which is applied to a master node in an OBTN, and includes:
- Time slot configuration module configured to implement the mapping process from the bandwidth authorization matrix to the bandwidth map. Specifically, it is configured to determine a location of an authorized time slot of a service sent by each source node to a different destination node in a data frame;
- the bandwidth map conversion module is configured to convert the time slot allocation result obtained by the time slot configuration module into a bandwidth map and send it to each slave node in the local optical burst transmission ring network.
- the time slot configuration method configured in sequence requires that the transmission slots of each source node be configured in order, and the source node configuration order is consistent with the OBTN running direction.
- the transmission slot configuration for each source node includes the operations of the slot prioritization unit and the slot configuration unit.
- the time slot configuration module may include:
- the time slot prioritization unit is configured to determine a selection priority of a time slot that the current source node sends to different destination nodes.
- the ordering principle of the selection priority of the time slot is: when each node with the OB having the current configuration has the smallest hop count (that is, the closest distance) as the destination node. The gap selection has the highest priority. In particular, If more than two time slots have the same number of hops that are not configured with the current configuration destination node, the priority of the time slots is equal;
- the time slot configuration unit is configured to implement an authorized time slot configuration of the current source node to different destination nodes. Specifically, for each source node, the transmission slot configuration for all the destination nodes is completed one by one according to the respective slot selection priorities of different destination nodes. For the slot configuration of each destination node, if there are more than two slots with the same priority, the order of slot selection is not fixed in principle. Generally, when the slot number is smaller, the slot number is preferred. Gap.
- the device may further include:
- a bandwidth authorization matrix generating module configured to obtain a bandwidth request matrix according to a bandwidth request of the entire network node, where an element in the bandwidth request matrix indicates a bandwidth request of the node i to the node j in the OBTN, where i and j are respectively a positive integer less than or equal to N, where N is the total number of nodes in the OBTN;
- the reduction authorization module is configured to perform the reduction authorization on the bandwidth request matrix according to the existing bandwidth resource to obtain a bandwidth authorization matrix, and implement a calculation process from the bandwidth request matrix to the bandwidth authorization matrix. Specifically, the bandwidth authorization size between any two nodes is determined according to the existing bandwidth resource and the actual bandwidth request information of each slave node.
- the cut authorization module may include:
- the row reduction unit is configured to initially determine the bandwidth authorization size of each source node to different destination nodes.
- each row of the bandwidth grant matrix represents the transmission bandwidth grant of the source node. Since the network node is only equipped with a fixed-wavelength optical transmitter carrying data services, only one OB can be accessed per time slot. Therefore, the source node has the largest transmission bandwidth. Line rate limit;
- the column reduction unit is configured to initially determine a bandwidth authorization size of each destination node receiving traffic from different source nodes.
- the columns of the bandwidth grant matrix represent the receive bandwidth grant of the corresponding destination node. Since the network node is only equipped with one wavelength tunable optical receiver, each time slot can only drop one OB, so the destination node is also received the most. Line rate limit.
- the upper triangular portion and the lower triangular portion of the bandwidth request matrix respectively correspond to bandwidth requests of the data frame that are not across the primary node service and across the primary node service. To prevent the starring of the non-cross-node service from being completely occupied by the service of the primary node, the maximum bandwidth grant calculation sub-unit needs to be introduced.
- the sub-unit is configured to determine the upper triangle in each column of the bandwidth authorization matrix.
- the strategy for determining the maximum bandwidth grant amount of the upper triangular portion and the lower triangular portion in each column of the maximum bandwidth grant amount calculation subunit is: for each column in the current bandwidth request matrix, the maximum line rate is in a loop length period The difference between the actual bandwidth grant amount of the lower triangular portion of the same column in the previously received bandwidth grant matrix, as the maximum bandwidth grant amount of the triangular portion of the column; and the maximum bandwidth grant amount of the lower triangular portion of each column is the largest The difference between the line rate and the bandwidth reservation preset for the column.
- the preset bandwidth reservation for each column can ensure at least the guaranteed bandwidth of all the connections on the corresponding column, and the preset bandwidth reservations for each column may be the same or different.
- the cut authorization module may further include: an authorization synthesizing unit configured to synthesize initial authorization results obtained after each connection is subjected to row reduction and column reduction, and finally determine a bandwidth authorization matrix.
- an authorization synthesizing unit configured to synthesize initial authorization results obtained after each connection is subjected to row reduction and column reduction, and finally determine a bandwidth authorization matrix.
- OBTN's bandwidth-based authorization based on centralized control faces the problem of bandwidth allocation of multiple sources and multiple sinks, and needs to consider both the transmission capability limitation and the reception capability limitation. Since the bandwidth request matrix is subjected to row reduction and column reduction, respectively, the same connection may obtain different authorization results.
- the authorization synthesizing unit may select the minimum value that is authorized by the row reduction and the column reduction. When the value of a connection is reduced and the minimum value of the column reduction authorization is performed, the row and the column may have a surplus.
- Increment that is, the total bandwidth of the row is less than the maximum line rate, and the total bandwidth of the upper triangle or the lower triangle of the column is less than the corresponding maximum bandwidth grant
- the connection between the surplus column and the surplus row can be connected.
- the connection after the bandwidth authorization is added cannot cause the line whose line exceeds the maximum line rate, and cannot cause the constraint of the column exceeding the maximum bandwidth authorization amount.
- the authorization mode is reduced without significantly reducing the total amount of bandwidth authorization matrix authorization.
- the block can dispense with the authorized synthesis unit. In this case, you need to: In the column reduction unit, the column reduction object is the bandwidth request matrix that has been cut off; or, in the row reduction unit, the row reduction object is the bandwidth request matrix that has been cut through the column, which is equivalent to the reduction authorization.
- the module performs row reduction and column reduction on the bandwidth request matrix to obtain a bandwidth authorization matrix.
- the main features of the dynamic bandwidth scheduling method for the optical burst transmission ring network proposed in this embodiment are: providing fairness guarantee and QoS guarantee by reducing the authorization step; providing service conflict resolution across the primary node through the slot configuration step and Channel space reuse.
- the optical burst transmission ring network includes five network nodes distributed clockwise (in fact, the number of network nodes can be set as needed, in this embodiment, five are taken as an example), wherein Nodel is The master node is responsible for centralized control management such as bandwidth allocation, and the remaining nodes are slave nodes.
- DBA Dynamic Bandwidth Allocation
- the master node collects the bandwidth request of the entire network node, and then allocates the bandwidth according to the relevant resources, updates and delivers the bandwidth map; the slave node transmits the bandwidth map according to the master node. Do the corresponding receiving and sending operations.
- DBA Dynamic Bandwidth Allocation
- the DBA period defines an update interval for the neighboring two dynamic bandwidth allocations of the master node.
- the DBA period may be equal to or smaller than the loop length period, which ensures that the master node can quickly respond to the dynamic bandwidth of the entire network node. request.
- OBTN uses a wavelength tuning mechanism of Fixed Transmitter Tunable Receiver (FTTR).
- FTTR Fixed Transmitter Tunable Receiver
- Each node in the ring network transmits data services on fixed data wavelengths, and tunes and receives local downlink services in any data wavelength.
- each The nodes are also equipped with a pair of fixed wavelength optical transceivers configured to transmit and receive control information on the control channel.
- the above data wavelength is divided into fixed-length OBs, time slots between different data wavelengths are kept synchronized, and OBs of all data wavelengths in a plurality of burst slots constitute one data frame.
- the bandwidth map carried in the control frame is matched.
- the indication is to indicate the OB attribution on each wavelength slot of the data frame, and the indication information should at least include the source node and the destination node information of the OB.
- the data frame may be delayed by a fiber delay line (FDL, Fiber Delay Line) to ensure that when When a data frame arrives at a node, the node has completed processing the corresponding control frame.
- the optimal delay time of the fiber delay line is the processing time of the node to the control frame.
- the control frame needs to undergo photoelectric conversion processing at the node, and at least includes: reading a bandwidth map, determining that the node transmits and receives the corresponding data frame; and writing the buffer amount of the local data to be sent to a specific position of the control frame, for reporting the bandwidth request.
- the node After completing the read/write processing of the control frame and the local data transmission and reception operation, the node sends the data frame and the control frame that is electrically and optically converted to the next hop node in the ring network.
- control frames and data frames can optionally arrive or leave the node in an asynchronous manner.
- the fiber delay line can be omitted at the node, but a reasonable offset time needs to be set for the control frame and sent before the data frame, so that for each node in the ring network, the control frame precedes the corresponding The data frame arrives and the processing operation of the corresponding control frame is completed before the data frame arrives.
- the primary node After the control frame and the data frame are looped back to the primary node, the primary node performs the DBA according to the latest network-wide node bandwidth request, and determines a new bandwidth map and delivers the bandwidth map to the entire network node.
- the application example Based on the operation mechanism of the optical burst transmission ring network, the application example provides a dynamic bandwidth scheduling method for the optical burst ring network. As shown in FIG. 5, the method includes:
- Step 401 The master node sends a bandwidth map by sending a control frame, where the bandwidth map includes at least an OB sending indication of each source node.
- each row of the bandwidth map respectively indicates a slot transmission schedule of each source node, and the OB information on the slot is a destination node corresponding to the source node to be sent.
- the master node sends a preset default bandwidth map.
- the default bandwidth map divides all the wavelength slots into the whole network interconnection node.
- Step 402 In the process of sending the bandwidth map, the transit node performs a corresponding service receiving operation according to the received bandwidth map. Meanwhile, the transit node also fills in the local bandwidth request to the designated position of the control frame, and the control frame is Delivered to the next hop node;
- the transit node performs corresponding service receiving operations according to the received bandwidth map, including: the route node synthesizes the cross-ring bandwidth map and the currently received bandwidth map to obtain a received map, and performs corresponding services according to the received map.
- Receive operation refers to a bandwidth map sent by the primary node in the long period of the previous ring, and the cross-ring bandwidth map overlaps with the current bandwidth map.
- each row element indicates the receiving of each destination node. In the case, if the map element RnnH is received, it indicates that the destination node n receives the OB sent by the source node j in the slot m.
- the method for synthesizing the received bandwidth map and the cross-ring bandwidth map is as shown in FIG. 6, which includes: assuming that the element AnnH in each received bandwidth map indicates that the source node n sends the OB to the slot m.
- the destination node i when synthesizing the received bandwidth map, scans the cross-ring bandwidth map row by row or column by column, for each element in the cross-ring bandwidth map, if the value of the element is smaller than the row of the element No., the row number of the element is used as the value of the element in the receiving bandwidth map with the value of the element and the column number as the row number and the column number respectively; on the other hand, scanning this line by row or column by column a received bandwidth map, for each element in the bandwidth map, if the value of the element is greater than the line number of the element, the line number of the element is used as the receiving bandwidth map with the element
- the value and column number are used as the values of the elements of the row number and the column number, respectively.
- Step 403 The control frame including the bandwidth map is returned to the primary node after one round of looping, and the primary node parses and collects the bandwidth request of the entire network node, performs a reduction authorization step, and formulates a bandwidth authorization matrix.
- the reduction authorization may include three sub-steps of row reduction, column reduction, and authorization synthesis;
- the line reduction is to meet the transmission capability limit of each source node;
- the column reduction is to meet the receiving capacity limit of each destination node
- the authorization synthesis is to maximize the bandwidth authorization in the case where the connection between any two nodes satisfies the corresponding source node transmission capability limit and the corresponding destination node reception capability limitation.
- the authorization synthesis sub-step may be omitted without significantly reducing the total amount of bandwidth authorization.
- Step 404 After the bandwidth authorization matrix is formulated, the master node maps the bandwidth authorization matrix into a bandwidth map by using a time slot configuration, that is, configuring the time slot authorization of each source node to different destination nodes in the bandwidth map by using the time slot configuration.
- a time slot configuration that is, configuring the time slot authorization of each source node to different destination nodes in the bandwidth map by using the time slot configuration.
- the time slot configuration step requires implementation of collision free transmission and reception of data packets.
- step 404 the process returns to step 401.
- the time slot configuration order of each source node is consistent with the OBTN running direction, and the time slot transmission of the primary node is first configured.
- the configuration sequence of each destination node in each source node is also consistent with the OBTN running direction.
- the destination node with the smallest number of hops of the source node is preferentially configured, and the next destination node is started after all authorization configurations of the destination node are completed. Gap configuration.
- the time slot configuration step includes a time slot prioritization sub-step and a time slot configuration sub-step.
- the slot priority ordering is to determine the slot selection priority sent to different destination nodes under each source node.
- the time slot configuration of different destination nodes is preferred to those slots that have the smallest number of hops with the current configuration destination node and that have not yet been dropped.
- time slots of different destination nodes determined according to the prioritization of the time slots are selected as priority levels, and the current source nodes are configured with appropriate time slots for the authorization of different destination nodes, and finally a bandwidth map is obtained.
- the embodiment provides a dynamic bandwidth scheduling device for the optical burst transmission ring network, as shown in FIG. 7, the dynamic bandwidth scheduling device of the optical burst transmission ring network Includes:
- the cut authorization module 701 is configured to perform the bandwidth request matrix according to an existing bandwidth resource. Reduce the authorization and get the bandwidth authorization matrix;
- the time slot configuration module 702 is configured to: for each source node in the OBTN, when allocating time slots for the connection from the source node to the destination node, in each time slot of the burst time slot OB where there is no downlink, priority The time slot occupied by the destination node with the smallest hop count of the currently configured destination node is selected; the bandwidth request matrix generation module 703 is configured to obtain a bandwidth request matrix according to the bandwidth request of the node of the entire network;
- the bandwidth map conversion module 704 is configured to convert the time slot assignment result obtained by the time slot configuration module into a bandwidth map and send it to each slave node in the OBTN.
- FIG. 8 is a schematic diagram of a cut authorization module 701, including: a row reduction unit 7011, a column reduction unit 7012, and an authorization synthesis unit 7013;
- a line reduction unit 7011 configured to perform line cut on the bandwidth request matrix
- Column reduction unit 7012 configured to perform column reduction on the matrix obtained after the row is cut.
- the authorization synthesizing unit 7013 is configured to: when determining the bandwidth authorization matrix, take the smaller one of the values obtained by the row reduction and the same position of the matrix obtained after the column reduction, as the bandwidth authorization. The value of the element at the corresponding position in the matrix;
- Line reduction refers to the reduction and authorization of each line of the bandwidth request matrix to ensure that the total transmission bandwidth of each source node does not exceed the maximum line rate limit.
- the transmission capability limit value of each source node is the maximum line rate.
- the line reduction process provided by the embodiment of the present invention is as shown in FIG. 9a, and includes the following steps:
- Step 901a scanning the bandwidth request matrix progressively.
- Step 902a determining whether the sum of requests is greater than a maximum line rate, and if so, executing step 903a; otherwise, executing step 906a.
- Step 903a reallocating the connection bandwidth.
- step 904a it is determined whether the remaining bandwidth is greater than 0. If yes, step 905a is performed; otherwise, step 906a is performed. In step 905a, the connection with the least saturation is selected, and the associated authorization is added to 1 OB.
- Step 906a completing the cutting authorization of the current scanning line.
- step 907a it is judged whether all of them have been scanned, and if so, step 908a is executed; otherwise, the process returns to step 901a.
- Step 908a completing the line cutting authorization.
- the types of bandwidth grants allocated for each connection on the line when the row is cut include: fixed bandwidth, guaranteed bandwidth, and non-guaranteed bandwidth.
- the allocation weight of the non-guaranteed bandwidth may be a service level agreement (SLA) or an actual request size.
- the minimum authorized unit of the bandwidth grant is 1 OB
- the bandwidth grant value based on the allocated weight is a non-integer
- the bandwidth grant based on the allocated weight is authorized.
- the value is rounded down.
- the remaining bandwidth polling is assigned to the unsaturated connection on the line, and one OB is allocated for each poll, where is not saturated.
- a connection is a connection that has an authorized bandwidth that is less than the actual requested bandwidth.
- Column reduction refers to the reduction authorization of each column of the bandwidth request matrix to ensure that the total time slot reception of each destination node does not exceed the maximum line rate limit.
- the column reduction process provided in this example is shown in Figure 9b and includes:
- Step 901b scanning the bandwidth request matrix column by column.
- Step 902b determining whether it is satisfied: the sum of the upper triangular portion of the current column is greater than the maximum bandwidth grant amount; or, the sum of the lower triangular portion of the current column is greater than the maximum bandwidth grant amount; if yes, step 903b is performed; otherwise, step 904b is performed. .
- Step 903b assigning a fixed bandwidth and a guaranteed bandwidth to each connection, and executing step 907b.
- Step 904b completing the crop authorization of the current scan column.
- step 905b it is determined whether all the columns have been scanned. If yes, step 906b is performed; otherwise, step 901b is returned. In step 906b, the column tailoring authorization is completed, and an authorization matrix is obtained.
- each time slot can only be tuned to receive burst data at a certain wavelength, that is, each node has a limit of maximum receiving capability. Therefore, each time a bandwidth grant matrix is formulated, it is necessary to consider the slot occupancy of the cross-ring frame that has not yet been routed.
- the cross-ring frame refers to a data frame corresponding to the previous ring long period overlapping with the current bandwidth map.
- the analysis shows that the service reception of each node in the ring network includes the non-cross-node service of the current frame and the cross-master service corresponding to the cross-ring frame.
- the upper triangular portion and the lower triangular portion of each column of the bandwidth authorization matrix respectively represent a non-main node service area and a cross-master node service area.
- the column clipping unit 7012 may further include a maximum The bandwidth grant amount calculation sub-unit 7014 is configured to obtain a maximum bandwidth grant amount of the triangular portion and the lower triangular portion of each column in the bandwidth grant matrix when the current bandwidth grant matrix is configured.
- the type of bandwidth grant assigned to each connection on the column when the column is cut includes fixed bandwidth, guaranteed bandwidth, and non-guaranteed bandwidth.
- the allocation weight of the non-guaranteed bandwidth can be SLA or the actual request size.
- the minimum authorization unit of the bandwidth grant in the example is 1 OB
- the bandwidth grant value based on the allocation weight allocation is a non-integer
- the allocation based on the allocation weight is The bandwidth grant value is rounded down.
- the bandwidth request matrix has met the transmission capability limit of the corresponding source node of each row and the reception capability limitation of each column destination node.
- the authorization synthesizing unit 7013 is required to synthesize the row authorization matrix obtained by the row clipping and the column authorization matrix obtained by the column clipping.
- Step 1101 Scan the row authorization matrix and the column authorization matrix row by row or column by column, and the corresponding connection bandwidth authorization should take the minimum authorization value in the row authorization matrix and the column authorization matrix.
- the bandwidth authorization of the connection is made to satisfy both the transmission capability limit of the row and the reception capability limit of the column.
- step 1102 a temporary authorization matrix (authized_mapl) is obtained, and step 1103a and step 1103b are performed.
- step 1103a the temporary authorization matrix is compared with the authorization row matrix to obtain a surplus vector of each row.
- step 1103b the temporary authorization matrix is differentiated from the authorized column matrix to obtain a surplus vector of each column, and the processing ends.
- Step 1103a and step 1103b can be performed in parallel.
- Step 1104 scanning the row heads of the authorization matrix row by line.
- step 1105 it is determined whether the obtained line surplus is greater than 0. If yes, step 1106 is performed; otherwise, step 1104 is returned.
- Step 1106 Scan the column surplus amount corresponding to each connection of the current row column by column.
- step 1107 it is determined whether each column on the current line has been scanned. If yes, the process returns to step 1104; otherwise, step 1108 is performed.
- Step 1108 Determine whether the column surplus is greater than 0. If yes, execute step 1109; otherwise, return to step 1106.
- Step 1109 adding a connection corresponding to an interaction point of a row and a column with a bandwidth surplus at the same time Add bandwidth grant, the increase is the minimum of the line surplus and the column surplus.
- the initially obtained bandwidth authorization matrix necessarily satisfies both the transmission capability limitation of the source node and the reception capability limitation of the destination node.
- the same connection since the same connection only takes the minimum authorized value of the row authorization matrix and the column authorization matrix, there may be a surplus bandwidth for each row and column in the bandwidth authorization matrix.
- each connection of the bandwidth authorization matrix may be scanned row by row and column by column. If the row and column of a connection have surplus bandwidth, the bandwidth authorization is added for the connection, wherein the maximum supplement may be The minimum amount of surplus between the row in which the connection is made and the column in which it is located.
- the surplus amount of the column in which a connection is located is the surplus amount of the upper triangular portion or the lower triangular portion of the connection.
- the cut authorization module 701 sets the authorization synthesizing unit 7013, the row cut and the column cut do not distinguish the order, and the roles of the two are the current bandwidth request matrix;
- the cut authorization module 701 may omit the set authorization synthesizing unit 7013 without significantly reducing the total amount of bandwidth authorization.
- the object to be cut is a column-reduced bandwidth request matrix, or an equal, column.
- the object of the cut is the bandwidth request matrix that has been rounded down.
- FIG. 12 is a schematic diagram of a time slot configuration module 702, including: a time slot prioritization unit 7021 and a time slot configuration unit 7022.
- the slot priority ordering unit 7021 is configured to determine a slot selection priority that each source node sends to different destination nodes. Specifically, the time slot of each destination node under each source node is configured in each time slot of the OB that has not been dropped, and the node with the smallest hop count of the currently configured destination node is preferentially selected as the time slot occupied by the destination node. . The purpose of this is to maximize the candidate slots that do not occupy other destination nodes to be configured.
- the time slot selects the priority.
- FIG. 13 is taken as an example.
- the transmission slot of the source node A is configured, the distribution of the data frame without the downlink OB is as shown in FIG. 13a.
- the time slot configuration method configured in sequence requires that the time slot configuration sent by the source node A to the destination node B preferentially selects the time slot with the smallest B hop count and the downlink OB, that is, the priority selected time slot exists.
- the time slot of the destination node C that is not the downlink OB is selected to be the second smallest hop to the current configuration destination node B.
- the destination node D does not have the time slot of the OB, as in the time slot ⁇ 7 ⁇ in the example.
- different priority levels may be set according to the hop count size, and the time slot on the time slot having the smallest hop count of the configuration destination node has the highest priority.
- the time slots ⁇ 1, 2, 6 ⁇ belong to the first selection priority level
- the time slot ⁇ 7 ⁇ belongs to the second selection priority level.
- the time slot should belong to the minimum hop count OB corresponding to the current configuration destination node.
- Priority level but within the priority level, the selection order of the time slots is delayed, as shown in FIG. 13a, the time slot configuration of the source node A to the destination node B, and the time slots ⁇ 1, 2, 6 ⁇ belong to the first
- the priority level is selected, but in the sixth time slot, in addition to the OB with the smallest number of hops of the node B and not sent to the node C, there is a OB that is not sent to the node D, so the time slot 6 is at the first.
- the selection order within the selection priority is later.
- the time slot is unavailable for the current configuration destination node.
- the graphical representation of the time slot selection priority may be as shown in FIG. 13b: each row represents a different destination node to which the current source node may be sent, and each column represents a slot number in the data frame, in particular, each row element is filled in.
- the numerical value indicates the priority of the time slot corresponding to the destination node. In this embodiment, the smaller the element value, the higher the priority.
- Fig. 13b "-1" indicates that the time slot is not available, and "kM+x" indicates that the priority of the time slot selection is the Xth selection order of the kth level.
- a slot configuration unit 7022 configured to each slot of the bandwidth grant matrix, slotted After the priority ranking is performed, the priority of the time slot selection sent by the current source node to different destination nodes is obtained. Subsequently, the time slot is selected according to the selection priority of the current source node to the service configuration of the different destination node.
- the time slot configuration of each destination node under the current source node preferentially selects the time slot with the highest priority.
- the current destination node selects the time slot of the next selected priority. Specifically, if all candidate slots whose traversal priority of the destination node to be configured are from high to low are used by other destination nodes, the authorization of the destination node to be configured is zeroed.
- the transmission slot configuration of the next source node is performed. After the transmission slots of all source nodes are configured, the bandwidth map update process ends and a new bandwidth map is obtained.
- the embodiment of the present invention further describes a dynamic bandwidth scheduling method, which is applied to a node in an OBTN, as shown in FIG. 14, and includes the following steps:
- Step 1401 When the target node is the master node, for each source node in the OBTN, when the target node allocates a time slot for the connection from the source node to the destination node, there is a burst slot OB that has not been dropped. In each time slot, the time slot occupied by the destination node with the smallest number of hops of the currently configured destination node is preferred.
- step 1402 when the target node is the master node, the time slot assignment result is converted into a bandwidth map and sent to each slave node in the OBTN.
- bandwidth authorization is performed for the connection between each source node and the destination node according to the bandwidth request of the entire network node.
- the bandwidth authorization is performed for the connection between each source node and the destination node according to the bandwidth request of the node of the entire network, including:
- the master node obtains a bandwidth request matrix according to a bandwidth request of a node of the entire network;
- the elements in the bandwidth request matrix represent bandwidth requests of node i to node j in the OBTN, i and j are positive integers less than or equal to N, respectively, and N is the total number of nodes in the OBTN;
- the master node performs the reduction authorization on the bandwidth request matrix according to the existing bandwidth resource, and obtains a bandwidth authorization matrix
- the element in the bandwidth authorization matrix represents the authorized bandwidth of the node i to the node j in the OBTN, and i and j are positive integers less than or equal to N, respectively, where N is the total number of nodes in the OBTN;
- the sum of the authorized bandwidths of each row element is less than or equal to the maximum line rate; for each column in the bandwidth grant matrix, the sum of the authorized bandwidths of all elements in the upper triangular portion of the column, and before a loop length period The sum of authorized bandwidths of all elements in the lower triangular portion of the same column in the received bandwidth grant matrix, less than or equal to the maximum line rate.
- the primary node performs the reduction authorization on the bandwidth request matrix according to the existing bandwidth resource, including:
- the master node performs row clipping on the bandwidth request matrix; the master node performs column reduction on the matrix obtained after the row is reduced.
- the performing, by the primary node, performing row clipping on the bandwidth request matrix includes: determining, for each row in the bandwidth request matrix, whether a sum of elements in the row exceeds the maximum line rate; if Re-allocating the bandwidth of each element in the row until the sum of the elements in the row is less than or equal to the maximum line rate;
- the primary node performs column reduction on the matrix obtained after the row is cut, and includes: determining, for each row in the matrix obtained by the row reduction, determining the sum of each element in the lower triangular portion of the column Whether the difference between the maximum line rate and the reserved bandwidth reserved for the column is exceeded; if it is exceeded, the bandwidth allocation is performed on each element in the lower triangular portion of the column until the resulting lower triangle of the column The sum of the elements in the portion is less than or equal to the difference between the maximum line rate and the reserved bandwidth reserved for the column; on the other hand, determining each of the upper triangular portions of the column Whether the sum of the elements exceeds the maximum bandwidth grant amount of the upper triangular portion of the column; if exceeded, the bandwidth allocation is performed on each element in the upper triangular portion of the column until the obtained upper triangular portion of the column a sum of elements, which is less than or equal to a maximum bandwidth grant amount of the upper triangular portion of the column; wherein, a maximum bandwidth grant amount of the upper
- the master node performs column clipping on the bandwidth request matrix, including: determining, for each column in the bandwidth request matrix, determining whether a sum of each element in a lower triangular portion of the column exceeds a maximum The difference between the line rate and the reserved bandwidth reserved for the column; if exceeded, the bandwidth allocation is performed on each element in the column until the sum of the elements in the lower triangular portion of the column is obtained, Less than or equal to the difference between the maximum line rate and the reserved bandwidth reserved for the column; on the other hand, determining whether the sum of each element in the upper triangular portion of the column exceeds the upper triangular portion of the column The maximum bandwidth grant amount; if exceeded, the bandwidth allocation is performed on each element in the upper triangular portion of the column until the sum of the elements in the upper triangular portion of the obtained column is less than or equal to the upper triangle of the column a maximum bandwidth grant amount of the portion; wherein a maximum bandwidth grant amount of the upper triangular portion of the column is the maximum line rate and
- Determining, by the master node, the matrix obtained after the column reduction includes: determining, for each row in the matrix obtained by the column reduction, whether the sum of each element in the row exceeds the maximum line rate; If it is exceeded, the bandwidth allocation is performed on each element in the row until the sum of the elements in the obtained row is less than or equal to the maximum line rate.
- the performing, by the primary node, performing row clipping on the bandwidth request matrix includes: determining, for each row in the bandwidth request matrix, whether a sum of elements in the row exceeds the maximum line rate; if Re-allocating the bandwidth of each element in the row until the sum of the elements in the row is less than or equal to the maximum line rate;
- the master node performs column reduction on the bandwidth request matrix, including:
- the method further includes:
- the primary node After determining the value of each element in the bandwidth authorization matrix, the primary node scans the bandwidth authorization matrix row by row and column by column, for a row whose sum of elements in the row is less than the maximum line rate And increasing the value of the element at the intersection of the sum of the elements in the upper triangular portion of the column and the column of the maximum bandwidth grant amount smaller than the upper triangular portion of the column, and after the addition, each of the elements in the row The sum of the elements does not exceed the maximum line rate, the sum of the elements in the upper triangular portion of the column in which the element is located does not exceed the maximum bandwidth grant amount of the upper triangular portion of the column; for the intersection of the row and the column The value of the element is increased, the behavior includes a row of each element that is smaller than the maximum line rate, and the column satisfies: a sum of elements in a lower triangular portion of the bandwidth authorization matrix, less than the maximum Line rate and bandwidth reserved for the column a difference, after the addition, the sum of the
- the re-allocating the bandwidth of each element in the column includes: the highest priority according to the fixed bandwidth allocation mode, the second priority of the guaranteed bandwidth allocation mode, and the lowest priority of the non-guaranteed bandwidth allocation mode.
- the order of bandwidth allocation for each element in the column correspondingly with the highest priority bandwidth allocation currently available;
- Re-allocating the bandwidth of each element in the row including:
- the currently available highest priority bandwidth allocation manner is correspondingly Each element in the row is allocated bandwidth.
- the method further includes:
- bandwidth allocation is performed on the elements in the column or the row by using a non-guaranteed bandwidth allocation manner, if there is still remaining bandwidth in the column or row, the remaining bandwidth polling is correspondingly allocated to the column or row.
- one OB is allocated per poll; where the unsaturation is a connection with an existing bandwidth grant less than the actual bandwidth request.
- the method further includes:
- the master node When the time slot allocation is performed, the master node performs time slot configuration in sequence from the master node in sequence according to the node arrangement order in the OBTN ring;
- the order of configuration of each destination node in each of the source nodes is as follows: According to the order of nodes in the OBTN ring, time slot configuration is performed in order of the number of hops of each destination node from the source node in descending order.
- the method further includes:
- the time slot with the smallest slot number is preferred.
- the method further includes:
- the target node When the target node is a slave node, the target node synthesizes the received bandwidth map and the cross-ring bandwidth map and obtains a received map, and receives the corresponding data frame according to the received map;
- the cross-ring bandwidth map is a bandwidth map received by the target node before a loop length period; in the receiving map, the element RnnH indicates that the destination node 11 receives the OB sent by the source node j in the slot m; In the map, the element AnnH indicates that the source node n sends the OB to the destination node i in the time slot m; m, n, i, and j are all positive integers; wherein, the bandwidth map and the cross-ring bandwidth map received by the current time are Synthesize to get the received map, including:
- the method further includes:
- the node delays the received data frame by using a fiber delay line; wherein, the delay time is greater than or equal to the control node of the local node Processing time.
- the embodiment of the present invention further describes a dynamic bandwidth scheduling apparatus, which is applied to a main node in an OBTN, as shown in FIG. 15a, and includes:
- a time slot configuration module 151 configured to be the source for each source node in the OBTN
- the conversion module 152 is configured to convert the time slot allocation result obtained by the time slot configuration module 151 into a bandwidth map and send it to each slave node in the OBTN.
- the method further includes: a bandwidth authorization matrix generating module 153, configured to obtain a bandwidth request matrix according to a bandwidth request of the entire network node, where an element in the bandwidth request matrix indicates the The bandwidth request of the node i to the node j in the OBTN, i and j are respectively positive integers equal to or less than N, and N is the total number of nodes in the OBTN;
- the reduction authorization module 154 is configured to perform the reduction authorization on the bandwidth request matrix according to the existing bandwidth resource to obtain a bandwidth authorization matrix.
- the element in the bandwidth authorization matrix indicates the authorized bandwidth of the node i to the node j in the OBTN.
- i and j are positive integers less than or equal to N, respectively, N is the total number of nodes in the OBTN; in the bandwidth authorization matrix, the sum of the authorized bandwidths of each row element is less than or equal to the maximum line rate; For each column, the sum of the authorized bandwidths of all elements in the upper triangular portion of the column and the sum of the authorized bandwidths of all elements in the lower triangular portion of the same column in the bandwidth grant matrix received before a ring length period, less than or equal to The maximum line rate.
- the cut authorization module 154 may include: a line cut unit 1541 configured to perform line cut on the bandwidth request matrix;
- the column reduction unit 1542 is configured to perform column reduction on the matrix obtained after the row is cut.
- the line reduction unit 1541 may include (not shown): a first determining subunit, configured to determine, for each row in the bandwidth request matrix, whether a sum of each element in the row is Exceeding the maximum line rate; the first allocating subunit, coupled to the first determining subunit, configured to re-perform each element in the row when the determining result of the first determining subunit is exceeded Bandwidth allocation until the sum of the elements in the obtained row is less than or equal to the maximum line rate;
- the column clipping unit 1542 may include: a second determining subunit configured to determine, for each column in the matrix obtained after the row clipping, whether the sum of each element in the lower triangular portion of the column exceeds a maximum line The difference between the rate and the reserved bandwidth reserved for the column; the second allocation subunit, coupled to the second determining subunit, configured to re-determine when the determining result of the second determining subunit is exceeded Bandwid
- a third determining subunit configured to determine, for each column in the matrix obtained after the row is cut, whether a sum of elements in the upper triangular portion of the column exceeds a maximum bandwidth authorization of an upper triangular portion of the column
- a third allocation subunit coupled to the third determining subunit, configured to re-perform each element in the upper triangular part of the column when the judgment result of the third determining subunit is exceeded Bandwidth allocation, until the sum of the elements in the upper triangular portion of the obtained column is less than or equal to the maximum bandwidth grant amount of the upper triangular portion of the column;
- the maximum bandwidth grant amount of the upper triangular portion of the column is the difference between the maximum line rate and the actual bandwidth grant amount of the lower triangle portion of the same column in the bandwidth grant matrix received before one loop length period.
- the line reduction unit 1541 includes: a fourth determining subunit, configured to determine, for each row in the matrix obtained after the column reduction, whether a sum of each element in the row exceeds the maximum line rate And a fourth allocation subunit, coupled to the fourth judging subunit, configured to perform bandwidth allocation on each element in the row when the judgment result of the fourth judging subunit is exceeded, until the obtained The sum of the elements in the row is less than or equal to the maximum line rate; wherein
- the column clipping unit 1542 includes: a fifth determining subunit configured to determine, for each column in the bandwidth request matrix, whether a sum of elements in a lower triangular portion of the column exceeds a maximum line rate and is The difference between the preset bandwidth reservations of the column; the fifth allocation subunit, and the The fifth determining subunit is coupled, configured to perform bandwidth allocation on each element in the column when the judgment result of the third determining subunit is exceeded, until the obtained elements in the lower triangular part of the column are And a difference between the maximum line rate and a reserved bandwidth reserved for the column;
- a sixth determining subunit configured to determine, for each column in the bandwidth request matrix, whether a sum of elements in the upper triangular portion of the column exceeds a maximum bandwidth grant amount of an upper triangular portion of the column;
- an allocation subunit coupled to the sixth judging subunit, configured to perform bandwidth allocation on each element in the upper triangular part of the column when the judgment result of the fifth judging subunit is exceeded, until obtained
- the sum of the elements in the upper triangular portion of the column is less than or equal to the maximum bandwidth grant amount of the upper triangular portion of the column; wherein, the maximum bandwidth grant amount of the upper triangular portion of the column is the maximum line rate and The difference between the actual bandwidth grant amount of the lower triangular portion of the same column in the bandwidth grant matrix received before a loop long period.
- the reduction of the authorization module 154 may further comprise:
- the authorization synthesizing unit 1543 is configured to, when determining the bandwidth authorization matrix, take the smaller one of the values of the elements obtained by the row reduction and the elements in the matrix obtained after the column reduction, as the bandwidth authorization. The value of the element at the corresponding position in the matrix;
- the line clipping unit 1541 may include: a seventh determining subunit, configured to determine, for each row in the bandwidth request matrix, whether a sum of elements in the row exceeds the maximum line rate; And an allocation subunit, coupled to the seventh judging subunit, configured to perform bandwidth allocation on each element in the row when the judgment result of the seventh judging subunit is exceeded, until the obtained row is The sum of each element in the medium is less than or equal to the maximum line rate;
- the column clipping unit 1542 includes: an eighth determining subunit configured to determine, in each of the bandwidth request matrices, a lower triangular portion of the column Whether the sum of each element exceeds a difference between a maximum line rate and a reserved bandwidth reserved for the column; an eighth allocation subunit, and the Eight determining subunit coupling, configured to perform bandwidth allocation on each element in the lower triangular portion of the column when the judgment result of the eighth determining subunit is exceeded, until the obtained lower triangular portion of the column The sum of the elements in the middle is less than or
- a ninth determining subunit configured to determine, for each column in the bandwidth request matrix, whether a sum of elements in an upper triangular portion of the column exceeds a maximum bandwidth grant amount of an upper triangular portion of the column; And allocating a subunit, coupled to the ninth judging subunit, configured to perform bandwidth allocation on each element in the upper triangular part of the column when the judgment result of the ninth judging subunit is exceeded, until obtained The sum of the elements in the upper triangular portion of the column is less than or equal to the maximum bandwidth grant amount of the upper triangular portion of the column; wherein the maximum bandwidth grant amount of the upper triangular portion of the column is the maximum line rate and The difference between the actual bandwidth grant amount of the lower triangular portion of the same column in the bandwidth grant matrix received before a loop long period.
- the first allocation sub-unit to the ninth allocation sub-unit are further configured to perform bandwidth allocation on each element in the re-pair, in the following manner:
- the currently available highest priority bandwidth allocation mode is correspondingly applied to the row.
- the currently available highest priority bandwidth allocation method is used for each of the columns. Elements are allocated for bandwidth.
- the first allocation subunit to the ninth allocation subunit are further configured to: after bandwidth allocation of elements in a column by using a non-guaranteed bandwidth allocation manner, if there is still remaining bandwidth in the column, the remaining bandwidth round The query is assigned to an unsaturated connection on the column, and each poll is allocated 1 OB; wherein, the unsaturated connection is a connection with an existing bandwidth grant amount smaller than the actual bandwidth request amount; After bandwidth allocation is performed on the elements in a row using the non-guaranteed bandwidth allocation method, if there is still bandwidth remaining in the row, the remaining bandwidth polling is allocated to the unsaturated connection on the row, and each polling is allocated 1 OB; where the unsaturated connection is a connection with an existing bandwidth grant amount smaller than the actual bandwidth request amount.
- the authorization synthesizing unit 1543 may include:
- the scanning subunit is configured to scan the bandwidth authorization matrix row by row and column by column after determining the value of each element in the bandwidth authorization matrix;
- a first adding subunit coupled to the scanning subunit, configured to have a sum of elements smaller than the maximum line rate and a sum of each element in an upper triangular portion of the column in the row being smaller than an upper triangle of the column
- the value of the element at the intersection of the columns of the maximum bandwidth grant amount is increased, and after the addition, the sum of the elements in the row of the element does not exceed the maximum line rate, and the upper triangular portion of the column in which the element is located The sum of the elements in the sum does not exceed the maximum bandwidth grant amount of the upper triangular portion of the column;
- a second increasing subunit coupled to the eleventh scan subunit, configured to have a sum of each element in the row that is less than the maximum line rate and a sum of each element in a lower triangular portion of the column that is smaller than the maximum
- the value of the element at the intersection of the line rate and the column of the difference between the bandwidth reservations preset for the column is increased, and after the addition, the sum of the elements in the row of the element does not exceed the maximum line rate
- the sum of the elements in the lower triangular portion of the column in which the element is located does not exceed the difference between the maximum line rate and the reserved bandwidth reserved for the column.
- the time slot configuration module 151 is further configured to perform time slot allocation according to the node arrangement order in the OBTN ring, and sequentially perform the time slot configuration of each node as a source node in sequence from the local node;
- the configuration sequence of each destination node is determined as follows: According to the node arrangement order in the OBTN ring, the time slot configuration sequence is sequentially performed according to the hop count of each destination node from the source node.
- the time slot configuration module 151 is further configured to be a source node to some When a connection of a destination node allocates a time slot, if there are two or more time slots occupied by the destination node with the smallest number of hops of the currently configured destination node in the time slots of the 0B that have not yet been dropped, the priority is given. The time slot in which the slot number is the smallest is selected.
- modules and units described in the embodiments of the present invention may be implemented by a central processing unit (CPU) in a dynamic bandwidth scheduling, a digital signal processor (DSP), or a field programmable gate array ( FPGA, Field Programmable Gate Array) implementation.
- CPU central processing unit
- DSP digital signal processor
- FPGA Field Programmable Gate Array
- the embodiment of the invention further describes a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the dynamic bandwidth scheduling method described above.
- the foregoing steps include the steps of the foregoing method embodiments; and the foregoing storage medium includes: a mobile storage device, a random access memory (RAM), a read-only memory (ROM), a magnetic disk or an optical disk, and the like.
- RAM random access memory
- ROM read-only memory
- the above-described integrated unit of the present invention may be stored in a computer readable storage medium if it is implemented in the form of a software functional module and sold or used as a standalone product.
- the technical solution of the embodiments of the present invention may be embodied in the form of a software product in essence or in the form of a software product.
- the computer software product is stored in a storage medium and includes a plurality of instructions.
- a computer device (which may be a personal computer, server, or network device, etc.) is implemented to perform all or part of the methods described in various embodiments of the present invention.
- the foregoing storage medium includes: a mobile storage device, a RAM, a ROM, a magnetic disk, or an optical disk, and the like, which can store a program code.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Small-Scale Networks (AREA)
Abstract
本发明实施例公开了一种动态带宽调度方法、装置及计算机存储介质,所述方法应用于光突发传送环网(OBTN)中的每一个节点,包括:在目标节点作为主节点时,对于OBTN中每一个源节点,在为所述源节点到某一目的节点的连接分配时隙时,在存在尚未下路的OB的各时隙中,优先选用与当前配置的目的节点跳数最小的目的节点所占用的时隙;在所述目标节点作为主节点时,将时隙分配结果转化为带宽地图的形式发送给OBTN中各从节点。
Description
动态带宽调度方法、 装置及计算机存储介质 技术领域
本发明涉及网络通信技术领域, 特别涉及一种动态带宽调度方法、 装 置及计算机存储介质。 背景技术
随着通信业务的高速拓展, 促成越来越多的用户接入和使用层出不穷 的宽带服务。 与此同时, 呈***式增长的网际协议(IP )业务在通信网络中 所占的比重不断增加, 使得通信网络的流量特征发生了明显的变化, 流量 相对平稳的传统语音业务在通信业务中的主流地位逐渐被具有流量突发性 高的数据分组业务所取代, 极大的增加了网络对大容量和灵活带宽分配的 需求。 从业务接口和光收发技术的发展趋势来看, 未来光网络应能够动态 灵活地提供不同的传输速率、 不同带宽粒度的子波长级全光交换能力。
新型的光突发传送环网 ( OBTN, Optical Burst Transport ring-Network ) 不仅能够友好地适应数据业务的突发性, 而且能有效降低全光网络对光器 件的苛刻要求, 具有灵活的组网能力和充分的可行性。 OBTN 的网络拓朴 如图 1所示,其中数据信道由若干波长组成(其中波长分别为: λ 0、 λ 1、 -.. λ Ν, 其中 Ν为正整数), 用于承载光突发数据; 控制信道釆用独立波长 λ c, 用于承载带宽分配、 时隙同步等控制信息; 数据帧与控制帧在各网络节 点处单独处理, 网络节点根据控制信息对数据帧进行相应的业务接收、 发 送操作。 可选地, OBTN各网络节点釆用固定波长发射机、 可调谐波长接 收机。
OBTN 以突发(burst )作为最小的网络交换单元, 每个突发由控制分 组(BCP, Burst Control Packet )和突发数据分组(BDP, Burst Data Packet ) 两部分组成, 两者在物理信道上分离传输, 网络节点获取控制信息后无需
等待确认即可直接实现数据分组的全光交换或透明传输。 这种控制信道与 数据信道分离的方式不仅在很大程度上简化了突发数据的交换处理, 而且 规避了目前光緩存技术尚未成熟的缺点, 降低了网络节点的实现复杂度。
基于媒质共享的光突发传送环网, 其实现的关键在于具有高效的媒质 接入控制 ( MAC, Media Access Control )技术, 即动态资源调度机制。 目 前, 光突发传送环网普遍釆用令牌接入或是基于冲突检测的随机接入方式。 从技术层面而言, 两者均属于分布式的控制管理。但分布式的 MAC技术容 易引起突发数据分组的竟争冲突, 造成光突发环网的带宽资源浪费。 此外, 由于分布式的媒质接入技术还要额外添加适当的控制协议以提供公平性保 证, 优选地增加了网络节点对控制信息的处理时延。
事实上, 基于时隙划分的集中式 MAC技术可以更好地提高网络性能: 在环网中选择一个节点作为主节点, 其余节点作为从节点。 主节点负责控 制平面的主要功能, 主要包括: 收集全网节点带宽请求、 根据相关资源分 配带宽、 更新并下发带宽地图等; 从节点根据主节点下发的带宽地图做相 应的接收发送操作。 在集中控制方案中, 所有的带宽调度策略都由主节点 控制, 这样可以全局统筹网络资源、 最大限度的提高带宽利用率、 提供公 平性保证并减小冲突竟争。
由于 OBTN的环网特性, 主节点处的动态带宽资源调度需要考虑跨主 节点业务的冲突影响。 其中, 跨主节点业务是指从源节点上路传送至相应 目的节点下路之前, 需要经过主节点的数据分组; 而非跨主节点业务是指 从源节点上路传送至相应目的节点下路之前, 无需经过主节点的数据分组。 以图 2为例,在某时刻 Tl,从节点 Node5按照当前的带宽地图在时隙 m上 路发往目的节点 Node3 的数据业务, 经过 时间后, 包含带宽地图的控制 帧重新到达主节点 Nodel。 此时, 主节点根据新的带宽请求进行带宽分配。 那么, 包括最新带宽地图的控制帧从 Nodel下发后, Node2将可能根据最
新的带宽地图同样在时隙 m上路发往目的节点 Node3的数据业务。 然而, 此时数据帧时隙 m上由 Node5发往 Node3的数据业务仍未下路, 两者势必 会造成冲突。 即使 Node2在时隙 m并不上路发往 Node3的数据业务, 当控 制帧和数据帧到达 Node3时, Node3也会根据新的带宽地图进行业务接收, 同样有可能得不到在时隙 T1由 Node5发出的数据业务。 发明内容
本发明实施例提供一种动态带宽调度方法、 装置及计算机存储介质, 可以应用于光突发传送环网, 能够解决基于时隙划分、 集中式控制的光突 发传送环网动态资源调度冲突的问题。
本发明实施例的技术方案是这样实现的:
本发明实施例提供一种动态带宽调度方法, 应用于 OBTN中的节点, 包括:
在目标节点作为主节点时, 对于 OBTN中每一个源节点, 所述目标节 点在为所述源节点到目的节点的连接分配时隙时, 在存在尚未下路的突发 时隙 OB的各时隙中,优先选用与当前配置的目的节点跳数最小的目的节点 所占用的时隙;
在所述目标节点作为主节点时, 将时隙分配结果转化为带宽地图的形 式发送给所述 OBTN中的各从节点。
优选地, 在所述目标节点作为主节点时, 根据全网节点的带宽请求, 为各源节点到目的节点间的连接进行带宽授权。
优选地, 所述根据全网节点的带宽请求, 为各源节点到目的节点间的 连接进行带宽授权, 包括:
所述主节点根据全网节点的带宽请求, 得到带宽请求矩阵; 其中, 所 述带宽请求矩阵中的元素 表示所述 OBTN中节点 i对节点 j的带宽请求, i和 j分别为小于等于 N的正整数, N为所述 OBTN中的节点总数;
所述主节点根据已有带宽资源对所述带宽请求矩阵进行裁减授权, 得 到带宽授权矩阵;
其中, 所述带宽授权矩阵中的元素 表示所述 OBTN中节点 i对节点 j的授权带宽, i和 j分别为小于等于 N的正整数, N为所述 OBTN中的节 点总数; 所述带宽授权矩阵中, 每一行元素的授权带宽总和小于等于最大 线速率; 对于所述带宽授权矩阵中的每一列, 所述列的上三角部分中的所 有元素的授权带宽总和, 与在一个环长周期前接收到的带宽授权矩阵中同 一列的下三角部分中的所有元素的授权带宽总和, 小于等于所述最大线速 率。
优选地, 所述主节点根据已有带宽资源对所述带宽请求矩阵进行裁减 授权, 包括:
所述主节点对所述带宽请求矩阵进行行裁减; 所述主节点对经过行裁 减后得到的矩阵进行列裁减。
优选地, 所述主节点对所述带宽请求矩阵进行行裁减, 包括: 对于所 述带宽请求矩阵中的每一行, 判断所述行中各元素的和是否超过所述最大 线速率; 若超过, 则重新对所述行中的各元素进行带宽分配, 直至得到的 所述行中的各元素的和小于等于所述最大线速率;
所述主节点对经过行裁减后得到的矩阵进行列裁减, 包括: 对于所述 经过行裁减后得到的矩阵中的每一列, 一方面, 判断所述列的下三角部分 中的各元素的和是否超过最大线速率与为所述列预设的带宽预留量之差; 若超过, 则重新对所述列的下三角部分中的各元素进行带宽分配, 直至得 到的所述列的下三角部分中各元素的和, 小于等于所述最大线速率与为所 述列预设的带宽预留量之差; 另一方面, 判断所述列的上三角部分中的各 元素的和是否超过所述列的上三角部分的最大带宽授权量; 若超过, 则重 新对所述列的上三角部分中的各元素进行带宽分配, 直至得到的所述列中
上三角部分中各元素的和, 小于等于所述列的上三角部分的最大带宽授权 量; 其中, 所述列的上三角部分的最大带宽授权量为所述最大线速率, 与 在一个环长周期前接收到的带宽授权矩阵中的相同列的下三角部分的实际 带宽授权量之差。
优选地, 所述主节点对所述带宽请求矩阵进行列裁减, 包括: 对于所 述带宽请求矩阵中的每一列, 一方面, 判断所述列的下三角部分中的各元 素的和是否超过最大线速率与为所述列预设的带宽预留量之差; 若超过, 则重新对所述列中的各元素进行带宽分配, 直至得到的所述列的下三角部 分中各元素的和, 小于等于所述最大线速率与为所述列预设的带宽预留量 之差; 另一方面, 判断所述列的上三角部分中的各元素的和是否超过所述 列的上三角部分的最大带宽授权量; 若超过, 则重新对所述列的上三角部 分中的各元素进行带宽分配, 直至得到的所述列的上三角部分中各元素的 和, 小于等于所述列的上三角部分的最大带宽授权量; 其中, 所述列的上 三角部分的最大带宽授权量, 为所述最大线速率与在一个环长周期前接收 到的带宽授权矩阵中的相同列的下三角部分的实际带宽授权量之差;
所述主节点对经过列裁减后得到的矩阵进行行裁减, 包括: 对于所述 经过列裁减后得到的矩阵中的每一行, 判断所述行中各元素的和是否超过 所述最大线速率; 若超过, 则重新对所述行中的各元素进行带宽分配, 直 至得到的所述行中的各元素的和小于等于所述最大线速率。
优选地, 所述主节点对所述带宽请求矩阵进行行裁减, 包括: 对于所 述带宽请求矩阵中的每一行, 判断所述行中各元素的和是否超过所述最大 线速率; 若超过, 则重新对所述行中的各元素进行带宽分配, 直至得到的 所述行中的各元素的和小于等于所述最大线速率;
所述主节点对所述带宽请求矩阵进行列裁减, 包括:
对于所述带宽请求矩阵中的每一列, 一方面, 判断所述列的下三角部
分中的各元素的和是否超过最大线速率与为所述列预设的带宽预留量之 差; 若超过, 则重新对所述列的下三角部分中的各元素进行带宽分配, 直 至得到的所述列的下三角部分中各元素的和, 小于等于所述最大线速率与 为所述列预设的带宽预留量之差; 另一方面, 判断所述列的上三角部分中 的各元素的和是否超过所述列的上三角部分的最大带宽授权量; 若超过, 则重新对所述列的上三角部分中的各元素进行带宽分配, 直至得到的所述 列中上三角部分中各元素的和, 小于等于所述列的上三角部分的最大带宽 授权量; 其中, 所述列的上三角部分的最大带宽授权量, 为所述最大线速 率与在一个环长周期前接收到的带宽授权矩阵中的相同列的下三角部分的 实际带宽授权量之差;
在确定所述带宽授权矩阵时, 取经过行裁减后得到的矩阵与经过列裁 减后得到的矩阵中位置相同的元素的值中较小的一个, 作为所述带宽授权 矩阵中相应位置的元素的值。
优选地, 所述方法还包括:
在确定了所述带宽授权矩阵中每一个元素的取值之后, 所述主节点逐 行及逐列对所述带宽授权矩阵进行扫描, 对于行中各元素的和小于所述最 大线速率的行、 及列中上三角部分中的各元素的和小于所述列的上三角部 分的最大带宽授权量的列的交汇点处的元素的值进行增加, 在增加后, 所 述元素所在行中各元素的和不超过所述最大线速率, 所述元素所在列的上 三角部分中的各元素的和不超过所述列的上三角部分的最大带宽授权量; 对于行、 及列的交汇点处的元素的值进行增加, 所述行为所包括各元素的 和小于所述最大线速率的行, 所述列满足: 所述带宽授权矩阵中下三角部 分中的各元素的和, 小于所述最大线速率与为所述列预设的带宽预留量之 差, 在增加后, 所述元素所在行中各元素的和不超过所述最大线速率, 所 述元素所在列中下三角部分中的各元素的和不超过所述最大线速率与为所
述列预设的带宽预留量之差。
优选地, 所述重新对所述列中的各元素进行带宽分配, 包括: 按照固定带宽分配方式的优先级最高、 保证带宽分配方式的优先级次 之、 及非保证带宽分配方式的优先级最低的次序, 釆用当前可用的优先级 最高的带宽分配方式相应地对所述列中的各元素进行带宽分配;
所述重新对所述行中的各元素进行带宽分配, 包括:
按照固定带宽分配方式的优先级最高、 保证带宽分配方式的优先级次 之、 及非保证带宽分配方式的优先级最低的次序, 釆用当前可用的优先级 最高的带宽分配方式相应地对所述行中的各元素进行带宽分配。
优选地, 所述方法还包括:
在釆用非保证带宽分配方式对所述列或所述行中的元素进行带宽分配 后, 若所述列或行中仍有剩余带宽, 则对应将剩余带宽轮询分配至所述列 或行上未饱和的连接, 每次轮询分配 1个 OB; 其中, 未饱和的连接为既有 带宽授权量小于实际带宽请求量的连接。
优选地, 所述方法还包括:
在进行时隙分配时, 所述主节点根据 OBTN环中的节点排列顺序, 从 所述主节点开始将各节点依次作为源节点按序进行时隙配置;
每个所述源节点下各目的节点的配置顺序为: 根据 OBTN环中的节点 排列顺序, 按照各目的节点距离所述源节点的跳数由小到大的顺序依次进 行时隙配置。
优选地, 所述方法还包括:
在为所述源节点到某一目的节点的连接分配时隙时, 若在存在尚未下 路的 OB的各时隙中,同时存在 2个以上的与当前配置的目的节点跳数最小 的目的节点所占用的时隙, 则优先选用时隙序号最小的时隙。
优选地, 所述方法还包括:
在所述目标节点作为从节点时, 所述目标节点对本次接收到的带宽地 图与跨环带宽地图及进行合成得到接收地图, 并根据所述接收地图对相应 数据帧进行接收; 其中, 所述跨环带宽地图为所述目标节点在一个环长周 期前接收到的带宽地图; 在所述接收地图中, 元素 RnnH 表示目的节点 11 在时隙 m接收源节点 j发出的 OB; 在素数带宽地图中, 元素 AnnH表示源 节点 n在时隙 m发送 OB至目的节点 i; m、 n、 i及 j均为正整数; 其中, 所述对本次接收到的带宽地图与跨环带宽地图及进行合成得到接收地 图, 包括:
逐行或逐列扫描所述跨环带宽地图, 针对所述跨环带宽地图中的每一 个元素, 若所述元素的取值小于所述元素所在行号, 则将所述元素的行号 作为所述接收带宽地图中以所述元素的取值及列号对应作为行号和列号的 元素的取值;
逐行或逐列扫描本次接收到的带宽地图, 针对所述带宽地图中的每一 个元素, 若所述元素的取值大于所述元素所在行号, 则将所述元素的行号 作为接收带宽地图中以所述元素的取值及列号对应作为行号和列号的元素 的取值。
优选地, 所述方法还包括:
当控制帧和对应的数据帧同时到达所述 OBTN中的节点时, 所述节点 通过光纤延时线对接收到的所述数据帧进行延迟; 其中, 延迟时间大于等 于本节点对所述控制帧的处理时间。
本发明实施例还提供一种动态带宽调度装置, 应用于 OBTN中的主节 点上, 包括:
时隙配置模块, 配置为对于 OBTN中每一个源节点, 在为所述源节点 到目的节点的连接分配时隙时, 在存在尚未下路的突发时隙 OB 的各时隙 中, 优先选用与当前配置的目的节点跳数最小的目的节点所占用的时隙;
带宽地图转化模块, 配置为将所述时隙配置模块得到的时隙分配结果 转化为带宽地图的形式发送给所述 OBTN中各从节点。
优选地, 还包括:
带宽授权矩阵生成模块, 配置为根据全网节点的带宽请求, 得到带宽 请求矩阵; 其中, 所述带宽请求矩阵中的元素 表示所述 OBTN中节点 i 对节点 j的带宽请求, i和 j分别为小于等于 N的正整数, N为所述 OBTN 中的节点总数;
裁减授权模块, 配置为根据已有带宽资源对所述带宽请求矩阵进行裁 减授权, 得到带宽授权矩阵; 其中, 所述带宽授权矩阵中的元素 表示所 述 OBTN中节点 i对节点 j的授权带宽, i和 j分别为小于等于 N的正整数, N为所述 OBTN中的节点总数; 所述带宽授权矩阵中, 每一行元素的授权 带宽总和小于等于最大线速率; 对于所述带宽授权矩阵中的每一列, 所述 列的上三角部分中的所有元素的授权带宽总和与在一个环长周期前接收到 的带宽授权矩阵中同一列的下三角部分中的所有元素的授权带宽总和, 小 于等于所述最大线速率。
优选地, 所述裁减授权模块包括:
行裁减单元, 配置为对所述带宽请求矩阵进行行裁减;
列裁减单元, 配置为对经过行裁减后得到的矩阵进行列裁减。
优选地, 所述行裁减单元包括: 第一判断子单元, 配置为对于所述带 宽请求矩阵中的每一行, 判断所述行中各元素的和是否超过所述最大线速 率; 第一分配子单元; 配置为当所述第一判断子单元的判断结果为超过时, 重新对所述行中的各元素进行带宽分配, 直至得到的所述行中的各元素的 和小于等于所述最大线速率;
所述列裁减单元包括: 第二判断子单元, 配置为对于所述经过行裁减 后得到的矩阵中的每一列, 判断所述列的下三角部分中的各元素的和是否
超过最大线速率与为所述列预设的带宽预留量之差; 第二分配子单元, 配 置为当所述第二判断子单元的判断结果为超过时, 重新对所述列的下三角 部分中的各元素进行带宽分配, 直至得到的所述列的下三角部分中各元素 的和小于等于所述最大线速率与为所述列预设的带宽预留量之差;
第三判断子单元, 配置为对于所述经过行裁减后得到的矩阵中的每一 列, 判断所述列的上三角部分中的各元素的和是否超过所述列的上三角部 分的最大带宽授权量; 第三分配子单元, 配置为当所述第三判断子单元的 判断结果为超过时, 则重新对所述列的上三角部分中的各元素进行带宽分 配, 直至得到的所述列中上三角部分中各元素的和小于等于所述列的上三 角部分的最大带宽授权量;
其中, 所述列的上三角部分的最大带宽授权量为所述最大线速率与在 一个环长周期前接收到的带宽授权矩阵中的相同列的下三角部分的实际带 宽授权量之差。
优选地, 所述行裁减单元包括: 第四判断子单元, 配置为对于所述经 过列裁减后得到的矩阵中的每一行, 判断所述行中各元素的和是否超过所 述最大线速率; 第四分配子单元, 配置为当所述第四判断子单元的判断结 果为超过时, 重新对所述行中的各元素进行带宽分配, 直至得到的所述行 中的各元素的和小于等于所述最大线速率; 其中,
所述列裁减单元包括: 第五判断子单元, 配置为对于所述带宽请求矩 阵中的每一列, 判断所述列的下三角部分中的各元素的和是否超过最大线 速率与为所述列预设的带宽预留量之差; 第五分配子单元, 配置为当所述 第三判断子单元的判断结果为超过时, 重新对所述列中的各元素进行带宽 分配, 直至得到的所述列的下三角部分中各元素的和小于等于所述最大线 速率与为所述列预设的带宽预留量之差;
第六判断子单元, 配置为对于所述带宽请求矩阵中的每一列, 判断所
述列的上三角部分中的各元素的和是否超过所述列的上三角部分的最大带 宽授权量; 第六分配子单元, 配置为当所述第五判断子单元的判断结果为 超过时, 重新对所述列的上三角部分中的各元素进行带宽分配, 直至得到 的所述列的上三角部分中各元素的和小于等于所述列的上三角部分的最大 带宽授权量; 其中, 所述列的上三角部分的最大带宽授权量为所述最大线 速率与在一个环长周期前接收到的带宽授权矩阵中的相同列的下三角部分 的实际带宽授权量之差。
优选地, 所述裁减授权模块还包括:
授权合成单元, 配置为在确定所述带宽授权矩阵时, 取经过行裁减后 得到的矩阵与经过列裁减后得到的矩阵中位置相同的元素的值中较小的一 个, 作为所述带宽授权矩阵中相应位置的元素的值;
所述行裁减单元包括: 第七判断子单元, 配置为对于所述带宽请求矩 阵中的每一行, 判断所述行中各元素的和是否超过所述最大线速率; 第七 分配子单元, 配置为当所述第七判断子单元的判断结果为超过时, 重新对 所述行中的各元素进行带宽分配, 直至得到的所述行中的各元素的和小于 等于所述最大线速率;
所述列裁减单元包括: 第八判断子单元, 配置为对于所述带宽请求矩 阵中的每一列, 判断所述列的下三角部分中的各元素的和是否超过最大线 速率与为所述列预设的带宽预留量之差; 第八分配子单元, 配置为当所述 第八判断子单元的判断结果为超过时, 重新对所述列的下三角部分中的各 元素进行带宽分配, 直至得到的所述列的下三角部分中各元素的和小于等 于所述最大线速率与为所述列预设的带宽预留量之差;
第九判断子单元, 配置为对于所述带宽请求矩阵中的每一列, 判断所 述列的上三角部分中的各元素的和是否超过所述列的上三角部分的最大带 宽授权量; 第九分配子单元, 配置为当所述第九判断子单元的判断结果为
超过时, 重新对所述列的上三角部分中的各元素进行带宽分配, 直至得到 的所述列中上三角部分中各元素的和小于等于所述列的上三角部分的最大 带宽授权量; 其中, 所述列的上三角部分的最大带宽授权量为所述最大线 速率与在一个环长周期前接收到的带宽授权矩阵中的相同列的下三角部分 的实际带宽授权量之差。
优选地, 所述授权合成单元包括:
扫描子单元 配置为在确定了所述带宽授权矩阵中每一个元素的取值 之后, 逐行及逐列对所述带宽授权矩阵进行扫描;
第一增加子单元, 配置为对于行中各元素的和小于所述最大线速率的 行及列中上三角部分中的各元素的和小于所述列的上三角部分的最大带宽 授权量的列的交汇点处的元素的值进行增加, 在增加后, 所述元素所在行 中各元素的和不超过所述最大线速率, 所述元素所在列的上三角部分中的 各元素的和不超过所述列的上三角部分的最大带宽授权量;
第二增加子单元, 配置为对于行中各元素的和小于所述最大线速率的 行及列中下三角部分中的各元素的和小于所述最大线速率与为所述列预设 的带宽预留量之差的列的交汇点处的元素的值进行增加, 在增加后, 所述 元素所在行中各元素的和不超过所述最大线速率, 所述元素所在列中下三 角部分中的各元素的和不超过所述最大线速率与为所述列预设的带宽预留 量之差。
优选地, 所述时隙配置模块还配置为在进行时隙分配时, 根据 OBTN 环中的节点排列顺序, 从本节点开始将各节点依次作为源节点按序进行时 隙配置;还配置为确定每个源节点下各目的节点的配置顺序为:根据 OBTN 环中的节点排列顺序, 按照各目的节点距离所述源节点的跳数由小到大的 顺序依次进行时隙配置序。
优选地, 所述时隙配置模块还配置为在为源节点到某一目的节点的连
接分配时隙时,若在存在尚未下路的 OB的各时隙中, 同时存在 2个以上的 与当前配置的目的节点跳数最小的目的节点所占用的时隙, 则优先选用其 中时隙序号最小的时隙。
本发明实施例还提供一种计算机存储介质, 所述计算机存储介质中存 储有计算机可执行指令, 所述计算机可执行指令用于执行以上所述的动态 带宽调度方法。
釆用本发明后, 可以实现光突发传送环网高效且无冲突的动态资源调 度, 在不中断业务的情况下完全解决跨主节点业务的冲突问题。 本发明不 仅能公平合理地分配带宽资源并快速响应突发业务的带宽需求, 而且能实 现数据无冲突交换、信道空间重用、严格服务质量(QoS, Quality of Service ) 保证以及获得较高的带宽利用率。 附图说明
图 1是相关技术中光突发传送环网示意图;
图 2是相关技术中 5节点光突发传送环网拓朴结构图;
图 3是本发明实施例中光突发传送环网的动态带宽调度方法的流程图; 图 4是本发明实施例的数据帧与带宽地图的图形表示;
图 5是光突发环网动态带宽调度方法的流程图;
图 6是本发明实施例的接收带宽地图的合成流程示意图;
图 7是本发明实施例中光突发传送环网的动态带宽调度装置结构图; 图 8是本发明实施例中裁减授权模块示意图;
图 9a和图 9b分别是本发明实施例中行裁减流程和列裁减流程图; 图 10是本发明实施例中带宽授权矩阵中跨主节点业务与非跨主节点业 务的区域示意图;
图 11是本发明实施例中授权合成流程图;
图 12是本发明实施例中时隙配置模块示意图;
图 13a和图 13b分别是本发明实施例的时隙优先级排序示例说明及时 隙选取优先级图形表示;
图 14是本发明实施例中动态带宽调度方法的流程图;
图 15a至图 15d是本发明实施例中动态带宽调度装置的结构示意图。 具体实施方式
为使本发明的目的、 技术方案和优点更加清楚明白, 下文中将结合附 图对本发明的实施例进行详细说明。 需要说明的是, 在不冲突的情况下, 本申请中的实施例及实施例中的特征可以相互任意组合。
本发明实施例记载一种光突发传送环网的动态带宽调度方法, 应用于 OBTN中的每一个节点, 如图 3所示, 包括:
步骤 301, 在所述目标节点作为主节点时, 对于 OBTN中每一个源节 点, 在为所述源节点到某一目的节点的连接分配时隙时, 在存在尚未下路 的突发时隙(OB, Optical Burst )的各时隙中, 优先选用与当前配置的目的 节点跳数最小的目的节点所占用的时隙;
步骤 302, 在所述目标节点作为主节点时, 将时隙分配结果转化为带宽 地图的形式发送给所述 OBTN中各从节点。
此外, 在所述节点作为主节点时, 还根据全网节点的带宽请求, 为各 源节点到目的节点间的连接进行带宽授权。
作为一个实施方式, 所述根据全网节点的带宽请求, 为各源节点到目 的节点间的连接进行带宽授权, 可以通过以下步骤实现:
步骤 10: 主节点 (也即目标节点)利用从控制帧信息中获取到的各从 节点的带宽请求, 组合得到带宽请求矩阵。
其中, 所述带宽请求矩阵中的元素 表示所述 OBTN中节点 i对节点 j的带宽请求, i和 j分别为小于等于 N的正整数, N为所述 OBTN中的节 点总数。 则在所述带宽请求矩阵中, 第 i行元素表示所述 OBTN 中节点 i
对所述 OBTN中的各节点的带宽请求, 第 i列元素表示所述 OBTN中各节 点对节点 1的带宽请求;
步骤 20:主节点根据已有带宽资源对上述带宽请求矩阵进行裁减授权, 得到带宽授权矩阵。
带宽授权矩阵确定了所述 OBTN中任意两节点间的授权带宽的大小, 即授权的 OB数量。 其中, 裁减授权包括行裁减和列裁减, 行裁减为: 逐行 对带宽请求矩阵进行带宽约束, 当某行的带宽请求总量超出所述行最大线 速率时, 需要对所述行中的元素重新进行带宽分配, 使得经过行裁减后得 到的所述行中带宽请求总量小于等于最大线速率; 带宽请求矩阵可以划分 为上三角部分和下三角部分; 其中, 上三角部分为非跨主节点业务的带宽 请求, 下三角部分为跨主节点业务的带宽请求, 则相应地, 列裁减为: 对 于带宽请求矩阵中的每一列, 对所述列中上三角部分的带宽请求总量和下 三角部分的带宽请求总量分别进行带宽约束, 当所述列中上三角部分或下 三角部分的带宽请求总量超出最大带宽授权量时, 需要相应的对所述列中 所述部分中的各元素重新进行带宽分配, 使得经过列裁减后得到的所述列 中所述部分的带宽请求总量小于等于最大带宽授权量。 裁减授权的目的是 为了保证所述 OBTN中源节点的发送和目的节点的接收均不超过最大线速 率限制。 特别地, 在进行行裁减及列裁减时, 可釆用的带宽分配方式包括: 固定带宽分配、 保证带宽分配和非保证带宽分配的组合, 以提供充分的公 平性保证和 QoS保证。 其中, 对于带宽请求矩阵中的每一列, 所述列中的 下三角部分的最大带宽授权量为最大线速率与为所述列预设的带宽预留量 之差; 所述列的上三角部分的最大带宽授权量为最大线速率与在一个环长 周期前接收到的带宽授权矩阵中的相同列的下三角部分的实际带宽授权量 之差;
步骤 30: 主节点通过时隙配置方法, 将上述带宽授权矩阵映射为时隙
分配表, 即带宽地图。 带宽地图的形式为二维矩阵, 如图 4所示, 假设所 述二维矩阵中的某一元素 B 的值为 k, 则表示节点 m在时隙 n上有发往 节点 k的 OB。 时隙配置方法为 "按序配置" 方法, 其核心思想是: (1 )顺 序配置源节点, 根据 OBTN中的节点排列顺序, 从主节点开始依次将各节 点作为源节点按序进行时隙配置; (2 )顺序配置目的节点, 在针对每个源 节点进行时隙配置时, 按照所述 OBTN的运行方向, 根据各目的节点距离 所述源节点的跳数由小到大的顺序, 依次配置所述源节点发往相应目的节 点的 OB所占用的时隙; (3 )针对各目的节点的时隙选取,在存在尚未下路 的 OB的各时隙中,优先选用与当前配置的目的节点跳数最小的目的节点所 占用的时隙。
本发明实施例还记载一种光突发传送环网的动态带宽调度装置, 应用 于 OBTN中的主节点上, 包括:
时隙配置模块: 配置为实现由带宽授权矩阵到带宽地图的映射过程。 具体地, 配置为确定各源节点发往不同目的节点的业务的授权时隙在数据 帧中所处的位置;
带宽地图转化模块, 配置为将时隙配置模块得到的时隙分配结果转化 为带宽地图的形式发送给本光突发传送环网中各从节点。
特别地, 按序配置的时隙配置方法要求按序配置各源节点的发送时隙, 源节点配置顺序与 OBTN运行方向一致。 对于每个源节点的发送时隙配置 都要包含时隙优先级排序单元和时隙配置单元的操作。
作为一个实施方式, 时隙配置模块可以包括:
时隙优先级排序单元, 配置为确定当前源节点发往不同目的节点的业 务可占用的时隙的选择优先级。 其中, 时隙的选择优先级的排序原则是: 在存在尚未下路的 OB的各时隙中,与当前配置的目的节点的跳数最小(即 距离最近)的节点作为目的节点所占用的时隙的选择优先级最高。 特别地,
如果 2个以上的时隙同时存在与当前配置目的节点跳数相同的未下路 OB, 则这些时隙的选择优先级相等;
时隙配置单元, 配置为实现当前源节点发往不同目的节点的授权时隙 配置。 具体地, 针对每一源节点, 根据不同目的节点各自的时隙选择优先 级逐一完成针对所有目的节点的发送时隙配置。 对于每个目的节点的时隙 配置而言, 若存在 2个以上的选择优先级相等的时隙, 则原则上不固定时 隙选择的顺序, 一般地, 可优先选用时隙序号较小的时隙。
作为一个实施方式, 装置还可以包括:
带宽授权矩阵生成模块, 配置为根据全网节点的带宽请求, 得到带宽 请求矩阵; 其中, 所述带宽请求矩阵中的元素 表示所述 OBTN中节点 i 对节点 j的带宽请求, i和 j分别为小于等于 N的正整数, N为所述 OBTN 中的节点总数;
裁减授权模块, 配置为根据已有带宽资源对所述带宽请求矩阵进行裁 减授权, 得到带宽授权矩阵, 实现由带宽请求矩阵到带宽授权矩阵的计算 过程。 具体地, 根据已有带宽资源和各从节点的实际带宽请求信息确定任 意两节点间的带宽授权大小
作为一个实施方式, 裁减授权模块可以包括:
行裁减单元, 配置为初步确定各源节点发往不同目的节点的带宽授权 大小。 实际上, 带宽授权矩阵各行表示源节点的发送带宽授权, 由于网络 节点只装备一个承载数据业务的固定波长光发射机, 每个时隙只能上路一 个 OB, 因此, 源节点的发送带宽受到最大线速率的限制;
列裁减单元, 配置为初步确定各目的节点接收来自不同源节点的业务 的带宽授权大小。 实际上, 带宽授权矩阵各列表示对应目的节点的接收带 宽授权, 由于网络节点只装备一个波长可调谐光接收机, 每个时隙只能下 路一个 OB, 因此, 目的节点的接收也受到最大线速率限制。
特别地, 带宽请求矩阵的上三角部分和下三角部分分别对应数据帧的 非跨主节点业务和跨主节点业务的带宽请求。 为了防止跨主节点业务完全 占用目的节点的接收能力, 造成非跨主节点业务的饿死现象, 需要引入最 大带宽授权量计算子单元, 所述子单元配置为确定带宽授权矩阵各列中上 三角部分和下三角部分的最大带宽授权量。 优选地, 最大带宽授权量计算 子单元确定各列中上三角部分和下三角部分的最大带宽授权量的策略是: 针对当前带宽请求矩阵中的每一列, 将最大线速率与在一个环长周期前接 收到的带宽授权矩阵中的相同列的下三角部分的实际带宽授权量的差, 作 为所述列上三角部分的最大带宽授权量; 而各列的下三角部分的最大带宽 授权量为最大线速率与为所在列预设的带宽预留量的差。 通常, 为各列预 设的带宽预留量至少能保证对应列上所有连接各自的保证带宽, 且为各列 预设的带宽预留量可以相同, 也可以不同。
作为一个实施方式, 裁减授权模块还可以包括: 授权合成单元, 配置 为合成各连接分别经过行裁减和列裁减后所得到的初步授权结果, 最终确 定带宽授权矩阵。 实际上, OBTN基于集中式控制的带宽授权所面对的是 多源多宿的带宽分配问题, 需要同时考虑发送能力限制和接收能力限制。 由于带宽请求矩阵分别经过行裁减和列裁减后, 同一连接可能得到不同的 授权结果。 此时, 针对所述连接, 授权合成单元可选取经过行裁减及列裁 减得到授权的最小值, 当某连接取值行裁减和列裁减授权的最小值后, 其 所在行和所在列可能存在盈余增量(即所在行的带宽总量小于最大线速率, 且所在列的上三角部分或下三角部分的带宽总量小于对应的最大带宽授权 量), 可对盈余列与盈余行交汇点的连接增补带宽授权, 进行带宽授权增补 后的连接不能造成其所在行超出最大线速率的约束, 且不能造成其所在列 超出最大带宽授权量的约束。
可选地, 在不明显降低带宽授权矩阵授权总量的情况下, 裁减授权模
块可省去授权合成单元。 此时需要: 在列裁减单元中, 列裁减对象为已经 过行裁减的带宽请求矩阵; 或者, 在行裁减单元中, 行裁减的对象为已经 过列裁减的带宽请求矩阵, 即相当于裁减授权模块对带宽请求矩阵先后进 行行裁减和列裁减, 得到带宽授权矩阵。
综上所示, 本实施例提出的光突发传送环网动态带宽调度方法的主要 特点是: 通过裁减授权步骤提供公平性保证及 QoS保证; 通过时隙配置步 骤提供跨主节点业务冲突解决及信道空间重用。
为使本发明的技术方法及技术优点更加清楚, 下面将结合附图对本发 明技术方案做进一步的详细描述。 需要指出的是, 下述实施例不应成为对 本发明的限制。
如图 2所示,假设光突发传送环网包含顺时针分布的 5个网络节点(实 际上, 网络节点数可按需设定, 本实施例中以 5个为例), 其中, Nodel为 主节点, 负责带宽分配等集中式控制管理, 其余节点为从节点。 每个动态 带宽分配( DBA, Dynamic Bandwidth Allocation )周期内, 主节点收集全网 节点的带宽请求, 再根据相关资源分配带宽、 更新和下发带宽地图; 从节 点则根据主节点下发的带宽地图做相应的接收、 发送操作。 优选地, DBA 周期定义为主节点进行相邻两次动态带宽分配的更新间隔, 在本实施例中, DBA周期可以等于或小于环长周期, 保证了主节点能快速响应全网节点的 动态带宽请求。
OBTN 釆用固定发送可调接收 (FTTR, Fixed Transmitter Tunable Receiver )的波长调谐机制,环网中各节点在固定数据波长上发送数据业务, 在任意数据波长中调谐接收本地下路业务; 此外, 每个节点还装备一对固 定波长的光收发机, 配置为收发控制信道上的控制信息。 将上述数据波长 划分为定长的 OB, 不同数据波长间的时隙保持同步, 且若干个突发时隙内 所有数据波长的 OB构成一个数据帧。相应的,控制帧中携带的带宽地图配
置为指示数据帧各波长时隙上的 OB归属,指示信息至少应包括 OB的源节 点和目的节点信息。
优选地, 为了防止包含带宽地图的控制帧和包含数据业务的数据帧同 步到达环网中的每个节点, 可通过光纤延时线( FDL, Fiber Delay Line )对 数据帧进行延迟, 以保证当数据帧到达某节点时, 所述节点已完成对相应 控制帧的处理。 其中, 光纤延时线的最佳延迟时间为节点对控制帧的处理 时间。 控制帧在节点处需要经过光电转换处理, 至少包括: 读取带宽地图, 确定本节点对相应数据帧的收发操作; 将本地待发送数据的緩存量写入控 制帧的特定位置, 用来上报带宽请求。 本节点在完成对控制帧的读写处理 及本地数据收发操作后, 将数据帧和经过电光转换的控制帧同步发往环网 中的下一跳节点。
需要指出的是: 事实上, 控制帧与数据帧可选择异步方式到达或离开 节点。 此时, 在节点处可省去光纤延时线, 但需要对控制帧设置合理的偏 置时间并先于数据帧发送, 使得对于环网中的每个节点而言, 控制帧均先 于对应的数据帧到达并在数据帧到达之前已完成对对应的控制帧的处理操 作。
所述控制帧和数据帧绕环一周后再次回到主节点, 主节点根据最新的 全网节点带宽请求进行 DBA, 确定新的带宽地图并下发至全网节点。 下面 用应用示例对本发明进行进一步说明。
基于上述光突发传送环网的运行机制, 本应用示例提供了一种光突发 环网的动态带宽调度方法, 如图 5所示, 包括:
步骤 401 : 主节点通过发送控制帧下发带宽地图, 所述带宽地图至少包 含各源节点的 OB发送指示。
具体地, 在 FTTR机制下, 带宽地图各行分别表示每个源节点的时隙 发送安排, 时隙上的 OB信息即为对应源节点下所要发送业务的目的节点。
***刚启动时主节点下发预先设定的默认带宽地图, 一般地, 默认带宽地 图把所有波长时隙均分至全网互联节点连接;
步骤 402: 在带宽地图的下发过程中, 途经节点根据接收到的带宽地图 进行相应的业务接收操作; 同时, 途经节点还将本地带宽请求填写至控制 帧的指定位置, 并将所述控制帧下发至下一跳节点;
优选地, 途经节点根据接收到的带宽地图进行相应的业务接收操作, 包括: 途径节点对跨环带宽地图和本次接收到的带宽地图进行合成得到接 收地图, 根据所述接收地图进行相应的业务接收操作。 其中, 跨环带宽地 图是指上一环长周期内主节点下发的带宽地图, 所述跨环带宽地图与当前 带宽地图存在时隙重叠; 在接收地图中, 各行元素表示各目的节点的接收 情况, 若接收地图元素 RnnH, 则表示目的节点 n在时隙 m接收源节点 j发 出的 OB。 特别地, 对本次接收到的带宽地图和跨环带宽地图进行合成的方 法如图 6所示, 包括:假设每次接收到的带宽地图中元素 AnnH表示源节点 n在时隙 m发送 OB至目的节点 i, 合成接收带宽地图时, 一方面, 逐行或 逐列扫描跨环带宽地图, 针对所述跨环带宽地图中的每一个元素, 若所述 元素的取值小于所述元素所在行号, 则将所述元素的行号作为接收带宽地 图中以所述元素的取值及列号分别作为行号和列号的元素的取值; 另一方 面, 逐行或逐列扫描本次接收到的带宽地图, 针对所述带宽地图中的每一 个元素, 若所述元素的取值大于所述元素所在行号, 则将所述元素的行号 作为接收带宽地图中以所述元素的取值及列号分别作为行号和列号的元素 的取值。
步骤 403: 包含所述带宽地图的控制帧绕环一周后重新回到主节点, 主 节点解析并收集全网节点的带宽请求, 执行裁减授权步骤, 制定带宽授权 矩阵。
优选地, 裁减授权可包括行裁减、 列裁减以及授权合成三个子步骤;
首先, 行裁减是为了满足各源节点的发送能力限制;
其次, 列裁减是为了满足各目的节点的接收能力限制;
最后, 授权合成是为了任意两节点间连接在同时满足相应源节点发送 能力限制和相应目的节点接收能力限制的情况下, 其带宽授权的最大化。 可选地, 在不明显降低带宽授权总量的情况下, 授权合成子步骤可省略。
步骤 404: 带宽授权矩阵制定完毕后, 主节点通过时隙配置将带宽授权 矩阵映射为带宽地图的形式, 即: 通过时隙配置把各源节点发往不同目的 节点的时隙授权配置在带宽地图合适的时隙内, 特别地, 时隙配置步骤要 求实现数据分组的无冲突发送和接收。
步骤 404之后, 返回步骤 401。
优选地, 各源节点的时隙配置顺序与 OBTN运行方向一致, 并首先配 置主节点的时隙发送。 每个源节点下各目的节点的配置顺序亦与 OBTN运 行方向一致, 优先配置与源节点跳数最小的目的节点, 且针对所述目的节 点的所有授权配置完毕后才开始下一目的节点的时隙配置。
优选地, 时隙配置步骤包括时隙优先级排序子步骤和时隙配置子步骤。 首先, 时隙优先级排序是为了确定各源节点下, 发往不同目的节点的 时隙选择优先级。 其中, 不同目的节点的时隙配置, 优先选用那些含有与 当前配置目的节点跳数最小、 且尚未下路 OB的时隙。
最后, 按照所述时隙优先级排序所确定的不同目的节点的时隙选用优 先级, 为当前源节点发往不同目的节点的授权配置合适的时隙, 最终得到 带宽地图。
相应地, 基于上述光突发传送环网的运行机制, 本实施例提供一种光 突发传送环网的动态带宽调度装置, 如图 7 所示, 光突发传送环网的动态 带宽调度装置包括:
裁减授权模块 701,配置为根据已有带宽资源对所述带宽请求矩阵进行
裁减授权, 得到带宽授权矩阵;
时隙配置模块 702, 配置为对于 OBTN中每一个源节点, 在为所述源 节点到目的节点的连接分配时隙时,在存在尚未下路的突发时隙 OB的各时 隙中, 优先选用与当前配置的目的节点跳数最小的目的节点所占用的时隙; 带宽请求矩阵生成模块 703, 配置为根据全网节点的带宽请求, 得到带 宽请求矩阵;
带宽地图转化模块 704,配置为将所述时隙配置模块得到的时隙分配结 果转化为带宽地图的形式发送给所述 OBTN中各从节点。
图 8是裁减授权模块 701的示意图, 包括: 行裁减单元 7011、 列裁减 单元 7012和授权合成单元 7013; 其中,
行裁减单元 7011, 配置为对所述带宽请求矩阵进行行裁减;
列裁减单元 7012, 配置为对经过行裁减后得到的矩阵进行列裁减。 授权合成单元 7013, 配置为在确定所述带宽授权矩阵时, 取经过行裁 减后得到的矩阵与经过列裁减后得到的矩阵中位置相同的元素的值中较小 的一个, 作为所述带宽授权矩阵中相应位置的元素的值;
行裁减是指对带宽请求矩阵各行进行裁减授权, 保证各源节点的发送 带宽总量不超出最大线速率限制, 特别地, 各源节点的发送能力限制值均 为最大线速率。 本发明实施例提供的行裁减流程如图 9a所示, 包括以下步 骤:
步骤 901a, 逐行扫描带宽请求矩阵。
步骤 902a, 判断请求总和是否大于最大线速率, 如果大于, 则执行步 骤 903a; 否则, 执行步骤 906a。
步骤 903a, 重新分配各连接带宽。
步骤 904a, 判断剩余带宽是否大于 0, 如果是, 则执行步骤 905a; 否 则, 执行步骤 906a。
步骤 905a, 选出饱和度最小的连接, 将其相关授权加 1个 OB。
步骤 906a, 完成当前扫描行的裁剪授权。
步骤 907a, 判断所有是否均已扫描, 如果是, 则执行步骤 908a; 否贝 'J, 返回步骤 901a。
步骤 908a, 完成行裁剪授权。
优选地, 行裁减时为行上各连接分配的带宽授权类型包括: 固定带宽、 保证带宽以及非保证带宽。 其中, 非保证带宽的分配权重可为服务等级协 定 ( SLA, Service-Level Agreement )或实际请求大小。
优选地, 由于本示例中, 带宽授权的最小授权单位为 1个 OB, 执行行 裁减非保证带宽分配时, 若基于分配权重分配的带宽授权值为非整数, 则 对基于分配权重分配的带宽授权值向下取整。 当行上所有连接均完成基于 分配权重的非保证带宽分配后, 若仍有剩余带宽, 则将剩余带宽轮询分配 至行上未饱和的连接, 每次轮询分配 1个 OB, 其中, 未饱和连接是指既有 授权带宽小于实际请求带宽的连接。
列裁减是指对带宽请求矩阵各列进行裁减授权, 保证各目的节点的时 隙接收总量不超出最大线速率限制。本示例提供的列裁减流程如图 9b所示, 包括:
步骤 901b, 逐列扫描带宽请求矩阵。
步骤 902b, 判断是否满足: 当前列的上三角部分请求总和大于最大带 宽授权量; 或者, 当前列的下三角部分请求总和大于最大带宽授权量; 如 果满足, 则执行步骤 903b; 否则, 执行步骤 904b。
步骤 903b, 为各连接分配固定带宽与保证带宽, 并执行步骤 907b。 步骤 904b, 完成当前扫描列的裁剪授权。
步骤 905b, 判断所有列是否均已扫描, 如果是, 则执行步骤 906b; 否 则, 返回步骤 901b。
步骤 906b, 完成列裁剪授权, 得到授权矩阵。
特别地, 由于环网中各节点基于固定发送可调接收的波长调谐机制, 每个时隙内只能调谐接收某一波长上的突发数据, 即每个节点都存在最大 接收能力的限制。 因此, 每次制定带宽授权矩阵时都需要考虑跨环帧中尚 未下路业务的时隙占用情况。 其中, 跨环帧是指与当前带宽地图重叠的上 一环长周期对应的数据帧。
分析表明, 环网中各节点的业务接收都包括当前帧的非跨主节点业务 和跨环帧对应的跨主节点业务。 如图 10所示, 带宽授权矩阵各列的上三角 部分和下三角部分分别表示非跨主节点业务区域和跨主节点业务区域。
优选地, 为了防止下一个跨环帧的非跨主节点业务出现 "饿死" 现象, 每次制定带宽授权矩阵时都应为下一个跨环帧各列的上三角部分提供带宽 预留, 同时, 把在一个环长周期前的带宽授权矩阵中各列的下三角部分未 用尽的带宽顺延给当前带宽授权矩阵相应列的上三角部分使用; 相应地, 列裁剪单元 7012中还可以包括最大带宽授权量计算子单元 7014,配置为制 定当前带宽授权矩阵时, 得到带宽授权矩阵中各列上三角部分和下三角部 分的最大带宽授权量。
优选地, 列裁减时为列上各连接分配的带宽授权类型包括固定带宽、 保证带宽以及非保证带宽。 其中, 非保证带宽的分配权重可为 SLA或实际 请求大小。
优选地, 由于本示例中的带宽授权的最小授权单位为 1个 OB, 执行所 述列裁减非保证带宽分配时, 若基于分配权重分配的带宽授权值为非整数, 则对基于分配权重分配的带宽授权值向下取整。 当列上所有连接均完成基 于分配权重的非保证带宽分配后, 若仍有剩余带宽, 则将剩余带宽轮询分 配至列上未饱和的连接, 每次轮询分配 1个 OB, 其中, 未饱和连接是指既 有授权带宽小于实际请求的连接。
分别经过行裁减和列裁减后, 带宽请求矩阵均已满足各行对应源节点 的发送能力限制和各列目的节点的接收能力限制。 然而, 由于光突发传送 环网为多源多宿的逻辑全互联, 每个连接的授权都会直接影响相应源节点 的发送能力限制和相应目的节点的接收能力限制。 因此, 需要授权合成单 元 7013 合成经由行裁减得到的行授权矩阵和经由列裁减得到的列授权矩 阵。 本示例提供的授权合成的流程如图 11所示, 包括:
步骤 1101, 逐行或逐列扫描行授权矩阵和列授权矩阵, 对应的同一连 接其带宽授权应取所述行授权矩阵和列授权矩阵中的最小授权值。
使得所述连接的带宽授权同时满足所在行的发送能力限制和所在列的 接收能力限制。
步骤 1102, 得到临时授权矩阵( authrized_mapl ), 执行步骤 1103a和 步骤 1103b。
步骤 1103a, 将临时授权矩阵与授权行矩阵作差, 得到各行盈余向量。 步骤 1103b, 将临时授权矩阵与授权列矩阵作差, 得到各列盈余向量, 结束处理。
步骤 1103a和步骤 1103b可以并行执行。
步骤 1104, 逐行扫描授权矩阵各行盈余量。
步骤 1105,判断得到的行盈余量是否大于 0,如果是,则执行步骤 1106; 否则, 返回步骤 1104。
步骤 1106, 逐列扫描当前行各连接对应的列盈余量。
步骤 1107,判断当前行上各列是否均已扫描,如果是,则返回步骤 1104; 否则, 执行步骤 1108。
步骤 1108, 判断列盈余量是否大于 0, 如果是, 则执行步骤 1109; 否 则, 返回步骤 1106。
步骤 1109, 对于同时存在带宽盈余的行和列的交互点所对应的连接增
加带宽授权, 增加量为行盈余量和列盈余量的最小值。
优选地, 经过上述授权合成过程, 初步得到的带宽授权矩阵必然同时 满足源节点的发送能力限制和目的节点的接收能力限制。 但是, 由于同一 连接仅取行授权矩阵和列授权矩阵的最小授权值, 此时的带宽授权矩阵中 各行和各列均可能存在盈余带宽。 为了充分利用带宽资源, 可逐行及逐列 扫描所述带宽授权矩阵的各个连接, 若某连接所在行和列均有盈余带宽, 则为所述连接增补带宽授权, 其中, 最大增补量可为所述连接所在行和所 在列两者间的最小盈余量。 特别地, 某连接所在列的盈余量是所述连接所 在上三角部分或下三角部分的盈余量。
优选地, 当裁减授权模块 701设置授权合成单元 7013时, 则行裁减和 列裁减不区分先后顺序, 两者的作用对象均为当前的带宽请求矩阵;
优选地, 在不明显降低带宽授权总量的情况下, 裁减授权模块 701 可 省略设置授权合成单元 7013。 需要指出的是, 为了严格保证各源节点的发 送能力限制和各目的节点的接收能力限制, 省略设置授权合成单元时, 要 求行裁减的对象是经过列裁减的带宽请求矩阵, 或平等地, 列裁减的对象 是经过行裁减的带宽请求矩阵。
图 12是时隙配置模块 702的示意图, 包括: 时隙优先级排序单元 7021 和时隙配置单元 7022。
时隙优先级排序单元 7021, 配置为确定各源节点发往不同目的节点的 时隙选择优先级。 特别地, 每个源节点下各目的节点的时隙配置在存在尚 未下路的 OB的各时隙中,优先选用与当前配置的目的节点的跳数最小的节 点作为目的节点所占用的时隙。 这样做的目的是为了最大可能的不占用其 他待配置目的节点的候选时隙。
优选地, 数据帧到达每个源节点时各时隙上都可能存在不同的未下路 OB的组合, 而这些未下路 OB的组合直接影响源节点发往不同目的节点的
时隙选择优先级。 具体地, 以图 13为例进行说明: 配置源节点 A的发送时 隙时,数据帧未下路 OB的分布情况如图 13a所示。按序配置的时隙配置方 法要求, 源节点 A发往目的节点 B的时隙配置优先选用时隙上存在与 B跳 数最小且尚未下路 OB的时隙, 即优先选取时隙上存在发往目的节点 C的 未下路 OB的时隙, 如图例中的时隙 { 1 ,2,6}; 其次, 再优先选取时隙上存在 与当前配置目的节点 B跳数第二小的发往目的节点 D未下路 OB的时隙, 如图例中的时隙 {7}。 特别地, 在时隙优先级排序时, 可按照跳数大小设置 不同的优先等级,时隙上存在与配置目的节点跳数最小的未下路 OB的时隙 其选择优先等级最高。 上述源节点 A发往目的节点 B的时隙选取中, 时隙 {1,2,6}属于第一选择优先等级, 时隙 {7}属于第二选择优先等级。
优选地, 若时隙上同时存在与当前配置目的节点跳数不同的 2个以上 的未下路 OB,则所述时隙应属于与当前配置目的节点间的跳数最小未下路 OB对应的优先等级, 但在所述优先等级内, 所述时隙的选取顺序延后, 如 图 13a中源节点 A发往目的节点 B的时隙配置, 时隙 {1,2,6}均属于第一选 择优先等级, 但 6号时隙除了存在与节点 B跳数最小且未下路发往节点 C 的 OB外,还存在发往节点 D的未下路 OB, 故 6号时隙在第一选择优先级 内的选取顺序靠后。
优选地, 若时隙上存在发往当前配置目的节点的未下路 OB, 则所述时 隙对于所述当前配置目的节点而言为不可用。
优选地, 所述时隙选取优先级的图形表示可为图 13b所示: 各行表示 当前源节点可能发往的不同目的节点, 各列表示数据帧中的时隙序号, 特 别地, 各行元素填写的数值表征对应目的节点的时隙选取优先级, 在本实 施例中, 元素数值越小表示优先级越高。 图 13b中: "-1 "表示时隙不可用, "kM+x"表示所述时隙选择的优先级为第 k等级的第 X个选择顺序。
时隙配置单元 7022, 配置为对于带宽授权矩阵的每个源节点, 时隙的
选择优先级排序执行完毕后, 得到当前源节点发往不同目的节点的时隙选 则优先级。 随后, 将按照所述选择优先级为当前源节点到不同目的节点的 业务配置发送时隙。
优选地, 当前源节点下每个目的节点的时隙配置优先选取选择优先级 最高的时隙。
优选地, 若候选时隙已被当前源节点的其他目的节点所使用, 则为当 前目的节点选取下一选择优先级的时隙。 特别地, 若当前待配置目的节点 遍历优先级从高到低的所有候选时隙均被其他目的节点所使用, 则将所述 待配置目的节点的授权置零。
特别地, 若当前源节点下的所有目的节点均按照时隙选择优先级完成 时隙配置后, 执行下一源节点的发送时隙配置。 在所有源节点的发送时隙 均配置完毕后, 带宽地图更新过程结束, 得到新的带宽地图。
本发明实施例还记载一种动态带宽调度方法,应用于 OBTN中的节点, 如图 14所示, 包括以下步骤:
步骤 1401, 在目标节点作为主节点时, 对于 OBTN中每一个源节点, 所述目标节点在为所述源节点到目的节点的连接分配时隙时, 在存在尚未 下路的突发时隙 OB的各时隙中,优先选用与当前配置的目的节点跳数最小 的目的节点所占用的时隙。
步骤 1402, 在所述目标节点作为主节点时, 将时隙分配结果转化为带 宽地图的形式发送给所述 OBTN中的各从节点。
优选地, 在所述目标节点作为主节点时, 根据全网节点的带宽请求, 为各源节点到目的节点间的连接进行带宽授权。
优选地, 所述根据全网节点的带宽请求, 为各源节点到目的节点间的 连接进行带宽授权, 包括:
所述主节点根据全网节点的带宽请求, 得到带宽请求矩阵; 其中, 所
述带宽请求矩阵中的元素 表示所述 OBTN中节点 i对节点 j的带宽请求, i和 j分别为小于等于 N的正整数, N为所述 OBTN中的节点总数;
所述主节点根据已有带宽资源对所述带宽请求矩阵进行裁减授权, 得 到带宽授权矩阵;
其中, 所述带宽授权矩阵中的元素 表示所述 OBTN中节点 i对节点 j的授权带宽, i和 j分别为小于等于 N的正整数, N为所述 OBTN中的节 点总数; 所述带宽授权矩阵中, 每一行元素的授权带宽总和小于等于最大 线速率; 对于所述带宽授权矩阵中的每一列, 所述列的上三角部分中的所 有元素的授权带宽总和, 与在一个环长周期前接收到的带宽授权矩阵中同 一列的下三角部分中的所有元素的授权带宽总和, 小于等于所述最大线速 率。
优选地, 所述主节点根据已有带宽资源对所述带宽请求矩阵进行裁减 授权, 包括:
所述主节点对所述带宽请求矩阵进行行裁减; 所述主节点对经过行裁 减后得到的矩阵进行列裁减。
优选地, 所述主节点对所述带宽请求矩阵进行行裁减, 包括: 对于所 述带宽请求矩阵中的每一行, 判断所述行中各元素的和是否超过所述最大 线速率; 若超过, 则重新对所述行中的各元素进行带宽分配, 直至得到的 所述行中的各元素的和小于等于所述最大线速率;
所述主节点对经过行裁减后得到的矩阵进行列裁减, 包括: 对于所述 经过行裁减后得到的矩阵中的每一列, 一方面, 判断所述列的下三角部分 中的各元素的和是否超过最大线速率与为所述列预设的带宽预留量之差; 若超过, 则重新对所述列的下三角部分中的各元素进行带宽分配, 直至得 到的所述列的下三角部分中各元素的和, 小于等于所述最大线速率与为所 述列预设的带宽预留量之差; 另一方面, 判断所述列的上三角部分中的各
元素的和是否超过所述列的上三角部分的最大带宽授权量; 若超过, 则重 新对所述列的上三角部分中的各元素进行带宽分配, 直至得到的所述列中 上三角部分中各元素的和, 小于等于所述列的上三角部分的最大带宽授权 量; 其中, 所述列的上三角部分的最大带宽授权量为所述最大线速率, 与 在一个环长周期前接收到的带宽授权矩阵中的相同列的下三角部分的实际 带宽授权量之差。
优选地, 所述主节点对所述带宽请求矩阵进行列裁减, 包括: 对于所 述带宽请求矩阵中的每一列, 一方面, 判断所述列的下三角部分中的各元 素的和是否超过最大线速率与为所述列预设的带宽预留量之差; 若超过, 则重新对所述列中的各元素进行带宽分配, 直至得到的所述列的下三角部 分中各元素的和, 小于等于所述最大线速率与为所述列预设的带宽预留量 之差; 另一方面, 判断所述列的上三角部分中的各元素的和是否超过所述 列的上三角部分的最大带宽授权量; 若超过, 则重新对所述列的上三角部 分中的各元素进行带宽分配, 直至得到的所述列的上三角部分中各元素的 和, 小于等于所述列的上三角部分的最大带宽授权量; 其中, 所述列的上 三角部分的最大带宽授权量, 为所述最大线速率与在一个环长周期前接收 到的带宽授权矩阵中的相同列的下三角部分的实际带宽授权量之差;
所述主节点对经过列裁减后得到的矩阵进行行裁减, 包括: 对于所述 经过列裁减后得到的矩阵中的每一行, 判断所述行中各元素的和是否超过 所述最大线速率; 若超过, 则重新对所述行中的各元素进行带宽分配, 直 至得到的所述行中的各元素的和小于等于所述最大线速率。
优选地, 所述主节点对所述带宽请求矩阵进行行裁减, 包括: 对于所 述带宽请求矩阵中的每一行, 判断所述行中各元素的和是否超过所述最大 线速率; 若超过, 则重新对所述行中的各元素进行带宽分配, 直至得到的 所述行中的各元素的和小于等于所述最大线速率;
所述主节点对所述带宽请求矩阵进行列裁减, 包括:
对于所述带宽请求矩阵中的每一列, 一方面, 判断所述列的下三角部 分中的各元素的和是否超过最大线速率与为所述列预设的带宽预留量之 差; 若超过, 则重新对所述列的下三角部分中的各元素进行带宽分配, 直 至得到的所述列的下三角部分中各元素的和, 小于等于所述最大线速率与 为所述列预设的带宽预留量之差; 另一方面, 判断所述列的上三角部分中 的各元素的和是否超过所述列的上三角部分的最大带宽授权量; 若超过, 则重新对所述列的上三角部分中的各元素进行带宽分配, 直至得到的所述 列中上三角部分中各元素的和, 小于等于所述列的上三角部分的最大带宽 授权量; 其中, 所述列的上三角部分的最大带宽授权量, 为所述最大线速 率与在一个环长周期前接收到的带宽授权矩阵中的相同列的下三角部分的 实际带宽授权量之差;
在确定所述带宽授权矩阵时, 取经过行裁减后得到的矩阵与经过列裁 减后得到的矩阵中位置相同的元素的值中较小的一个, 作为所述带宽授权 矩阵中相应位置的元素的值。
优选地, 所述方法还包括:
在确定了所述带宽授权矩阵中每一个元素的取值之后, 所述主节点逐 行及逐列对所述带宽授权矩阵进行扫描, 对于行中各元素的和小于所述最 大线速率的行、 及列中上三角部分中的各元素的和小于所述列的上三角部 分的最大带宽授权量的列的交汇点处的元素的值进行增加, 在增加后, 所 述元素所在行中各元素的和不超过所述最大线速率, 所述元素所在列的上 三角部分中的各元素的和不超过所述列的上三角部分的最大带宽授权量; 对于行、 及列的交汇点处的元素的值进行增加, 所述行为所包括各元素的 和小于所述最大线速率的行, 所述列满足: 所述带宽授权矩阵中下三角部 分中的各元素的和, 小于所述最大线速率与为所述列预设的带宽预留量之
差, 在增加后, 所述元素所在行中各元素的和不超过所述最大线速率, 所 述元素所在列中下三角部分中的各元素的和不超过所述最大线速率与为所 述列预设的带宽预留量之差。
优选地, 所述重新对所述列中的各元素进行带宽分配, 包括: 按照固定带宽分配方式的优先级最高、 保证带宽分配方式的优先级次 之、 及非保证带宽分配方式的优先级最低的次序, 釆用当前可用的优先级 最高的带宽分配方式相应地对所述列中的各元素进行带宽分配;
所述重新对所述行中的各元素进行带宽分配, 包括:
按照固定带宽分配方式的优先级最高、 保证带宽分配方式的优先级次 之、 及非保证带宽分配方式的优先级最低的次序, 釆用当前可用的优先级 最高的带宽分配方式相应地对所述行中的各元素进行带宽分配。
优选地, 所述方法还包括:
在釆用非保证带宽分配方式对所述列或所述行中的元素进行带宽分配 后, 若所述列或行中仍有剩余带宽, 则对应将剩余带宽轮询分配至所述列 或行上未饱和的连接, 每次轮询分配 1个 OB; 其中, 未饱和的连接为既有 带宽授权量小于实际带宽请求量的连接。
优选地, 所述方法还包括:
在进行时隙分配时, 所述主节点根据 OBTN环中的节点排列顺序, 从 所述主节点开始将各节点依次作为源节点按序进行时隙配置;
每个所述源节点下各目的节点的配置顺序为: 根据 OBTN环中的节点 排列顺序, 按照各目的节点距离所述源节点的跳数由小到大的顺序依次进 行时隙配置。
优选地, 所述方法还包括:
在为所述源节点到某一目的节点的连接分配时隙时, 若在存在尚未下 路的 OB的各时隙中,同时存在 2个以上的与当前配置的目的节点跳数最小
的目的节点所占用的时隙, 则优先选用时隙序号最小的时隙。
优选地, 所述方法还包括:
在所述目标节点作为从节点时, 所述目标节点对本次接收到的带宽地 图与跨环带宽地图及进行合成得到接收地图, 并根据所述接收地图对相应 数据帧进行接收; 其中, 所述跨环带宽地图为所述目标节点在一个环长周 期前接收到的带宽地图; 在所述接收地图中, 元素 RnnH 表示目的节点 11 在时隙 m接收源节点 j发出的 OB; 在素数带宽地图中, 元素 AnnH表示源 节点 n在时隙 m发送 OB至目的节点 i; m、 n、 i及 j均为正整数; 其中, 所述对本次接收到的带宽地图与跨环带宽地图及进行合成得到接收地 图, 包括:
逐行或逐列扫描所述跨环带宽地图, 针对所述跨环带宽地图中的每一 个元素, 若所述元素的取值小于所述元素所在行号, 则将所述元素的行号 作为所述接收带宽地图中以所述元素的取值及列号对应作为行号和列号的 元素的取值;
逐行或逐列扫描本次接收到的带宽地图, 针对所述带宽地图中的每一 个元素, 若所述元素的取值大于所述元素所在行号, 则将所述元素的行号 作为接收带宽地图中以所述元素的取值及列号对应作为行号和列号的元素 的取值。
优选地, 所述方法还包括:
当控制帧和对应的数据帧同时到达所述 OBTN中的节点时, 所述节点 通过光纤延时线对接收到的所述数据帧进行延迟; 其中, 延迟时间大于等 于本节点对所述控制帧的处理时间。
本发明实施例还记载一种动态带宽调度装置, 应用于 OBTN中的主节 点上, 如图 15a所示, 包括:
时隙配置模块 151, 配置为对于 OBTN中每一个源节点, 在为所述源
节点到目的节点的连接分配时隙时,在存在尚未下路的突发时隙 OB的各时 隙中, 优先选用与当前配置的目的节点跳数最小的目的节点所占用的时隙; 带宽地图转化模块 152,配置为将所述时隙配置模块 151得到的时隙分 配结果转化为带宽地图的形式发送给所述 OBTN中各从节点。
作为一个实施方式, 如图 15b所示, 还可以包括: 带宽授权矩阵生成 模块 153, 配置为根据全网节点的带宽请求, 得到带宽请求矩阵; 其中, 所 述带宽请求矩阵中的元素 表示所述 OBTN中节点 i对节点 j的带宽请求, i和 j分别为小于等于 N的正整数, N为所述 OBTN中的节点总数;
裁减授权模块 154,配置为根据已有带宽资源对所述带宽请求矩阵进行 裁减授权, 得到带宽授权矩阵; 其中, 所述带宽授权矩阵中的元素 表示 所述 OBTN中节点 i对节点 j的授权带宽, i和 j分别为小于等于 N的正整 数, N为所述 OBTN中的节点总数; 所述带宽授权矩阵中, 每一行元素的 授权带宽总和小于等于最大线速率; 对于所述带宽授权矩阵中的每一列, 所述列的上三角部分中的所有元素的授权带宽总和与在一个环长周期前接 收到的带宽授权矩阵中同一列的下三角部分中的所有元素的授权带宽总 和, 小于等于所述最大线速率。
作为一个实施方式, 如图 15c所示, 所述裁减授权模块 154可以包括: 行裁减单元 1541, 配置为对所述带宽请求矩阵进行行裁减;
列裁减单元 1542, 配置为对经过行裁减后得到的矩阵进行列裁减。 作为一个实施方式, 所述行裁减单元 1541 可以包括(图中未示出): 第一判断子单元, 配置为对于所述带宽请求矩阵中的每一行, 判断所述行 中各元素的和是否超过所述最大线速率; 第一分配子单元, 与所述第一判 断子单元耦合, 配置为当所述第一判断子单元的判断结果为超过时, 重新 对所述行中的各元素进行带宽分配, 直至得到的所述行中的各元素的和小 于等于所述最大线速率;
所述列裁减单元 1542可以包括: 第二判断子单元, 配置为对于所述经 过行裁减后得到的矩阵中的每一列, 判断所述列的下三角部分中的各元素 的和是否超过最大线速率与为所述列预设的带宽预留量之差; 第二分配子 单元, 与所述第二判断子单元耦合, 配置为当所述第二判断子单元的判断 结果为超过时, 重新对所述列的下三角部分中的各元素进行带宽分配, 直 至得到的所述列的下三角部分中各元素的和小于等于所述最大线速率与为 所述列预设的带宽预留量之差;
第三判断子单元, 配置为对于所述经过行裁减后得到的矩阵中的每一 列, 判断所述列的上三角部分中的各元素的和是否超过所述列的上三角部 分的最大带宽授权量; 第三分配子单元, 与所述第三判断子单元耦合, 配 置为当所述第三判断子单元的判断结果为超过时, 则重新对所述列的上三 角部分中的各元素进行带宽分配, 直至得到的所述列中上三角部分中各元 素的和小于等于所述列的上三角部分的最大带宽授权量;
其中, 所述列的上三角部分的最大带宽授权量为所述最大线速率与在 一个环长周期前接收到的带宽授权矩阵中的相同列的下三角部分的实际带 宽授权量之差。
优选地, 所述行裁减单元 1541包括: 第四判断子单元, 配置为对于所 述经过列裁减后得到的矩阵中的每一行, 判断所述行中各元素的和是否超 过所述最大线速率; 第四分配子单元, 与所述第四判断子单元耦合, 配置 为当所述第四判断子单元的判断结果为超过时, 重新对所述行中的各元素 进行带宽分配, 直至得到的所述行中的各元素的和小于等于所述最大线速 率; 其中,
所述列裁减单元 1542包括: 第五判断子单元, 配置为对于所述带宽请 求矩阵中的每一列, 判断所述列的下三角部分中的各元素的和是否超过最 大线速率与为所述列预设的带宽预留量之差; 第五分配子单元, 与所述第
五判断子单元耦合, 配置为当所述第三判断子单元的判断结果为超过时, 重新对所述列中的各元素进行带宽分配, 直至得到的所述列的下三角部分 中各元素的和小于等于所述最大线速率与为所述列预设的带宽预留量之 差;
第六判断子单元, 配置为对于所述带宽请求矩阵中的每一列, 判断所 述列的上三角部分中的各元素的和是否超过所述列的上三角部分的最大带 宽授权量; 第六分配子单元, 与所述第六判断子单元耦合, 配置为当所述 第五判断子单元的判断结果为超过时, 重新对所述列的上三角部分中的各 元素进行带宽分配, 直至得到的所述列的上三角部分中各元素的和小于等 于所述列的上三角部分的最大带宽授权量; 其中, 所述列的上三角部分的 最大带宽授权量为所述最大线速率与在一个环长周期前接收到的带宽授权 矩阵中的相同列的下三角部分的实际带宽授权量之差。
作为一个实施方式, 基于图 15C, 如图 15d所示, 所述裁减授权模块 154还可以包括:
授权合成单元 1543, 配置为在确定所述带宽授权矩阵时, 取经过行裁 减后得到的矩阵与经过列裁减后得到的矩阵中位置相同的元素的值中较小 的一个, 作为所述带宽授权矩阵中相应位置的元素的值;
相应地, 所述行裁减单元 1541可以包括: 第七判断子单元, 配置为对 于所述带宽请求矩阵中的每一行, 判断所述行中各元素的和是否超过所述 最大线速率; 第七分配子单元, 与所述第七判断子单元耦合, 配置为当所 述第七判断子单元的判断结果为超过时, 重新对所述行中的各元素进行带 宽分配, 直至得到的所述行中的各元素的和小于等于所述最大线速率; 所述列裁减单元 1542包括: 第八判断子单元, 配置为对于所述带宽请 求矩阵中的每一列, 判断所述列的下三角部分中的各元素的和是否超过最 大线速率与为所述列预设的带宽预留量之差; 第八分配子单元, 与所述第
八判断子单元耦合, 配置为当所述第八判断子单元的判断结果为超过时, 重新对所述列的下三角部分中的各元素进行带宽分配, 直至得到的所述列 的下三角部分中各元素的和小于等于所述最大线速率与为所述列预设的带 宽预留量之差;
第九判断子单元, 配置为对于所述带宽请求矩阵中的每一列, 判断所 述列的上三角部分中的各元素的和是否超过所述列的上三角部分的最大带 宽授权量; 第九分配子单元, 与所述第九判断子单元耦合, 配置为当所述 第九判断子单元的判断结果为超过时, 重新对所述列的上三角部分中的各 元素进行带宽分配, 直至得到的所述列中上三角部分中各元素的和小于等 于所述列的上三角部分的最大带宽授权量; 其中, 所述列的上三角部分的 最大带宽授权量为所述最大线速率与在一个环长周期前接收到的带宽授权 矩阵中的相同列的下三角部分的实际带宽授权量之差。
其中, 所述第一分配子单元至第九分配子单元, 还配置为对重新对行 中的各元素进行带宽分配时, 釆用以下方式:
按照固定带宽分配方式的优先级最高、 保证带宽分配方式的优先级次 之、 及非保证带宽分配方式的优先级最低的次序, 釆用当前可用的优先级 最高的带宽分配方式相应地对该行中的各元素进行带宽分配;
还配置为对列中的各元素进行带宽分配时, 釆用以下方式:
按照固定带宽分配方式的优先级最高、 保证带宽分配方式的优先级次 之、 及非保证带宽分配方式的优先级最低的次序, 釆用当前可用的优先级 最高的带宽分配方式对列中的各元素进行带宽分配。
所述第一分配子单元至第九分配子单元, 还配置为在釆用非保证带宽 分配方式对某列中的元素进行带宽分配后, 若该列中仍有剩余带宽, 则将 剩余带宽轮询分配至该列上未饱和的连接, 每次轮询分配 1个 OB; 其中, 未饱和的连接为既有带宽授权量小于实际带宽请求量的连接;
在釆用非保证带宽分配方式对某行中的元素进行带宽分配后, 若该行 中仍有剩余带宽, 则将剩余带宽轮询分配至该行上未饱和的连接, 每次轮 询分配 1个 OB; 其中, 未饱和的连接为既有带宽授权量小于实际带宽请求 量的连接。
作为一个实施方式, 所述授权合成单元 1543可以包括:
扫描子单元 配置为在确定了所述带宽授权矩阵中每一个元素的取值 之后, 逐行及逐列对所述带宽授权矩阵进行扫描;
第一增加子单元, 与所述扫描子单元耦合, 配置为对于行中各元素的 和小于所述最大线速率的行及列中上三角部分中的各元素的和小于所述列 的上三角部分的最大带宽授权量的列的交汇点处的元素的值进行增加, 在 增加后, 所述元素所在行中各元素的和不超过所述最大线速率, 所述元素 所在列的上三角部分中的各元素的和不超过所述列的上三角部分的最大带 宽授权量;
第二增加子单元, 与所述第十一扫描子单元耦合, 配置为对于行中各 元素的和小于所述最大线速率的行及列中下三角部分中的各元素的和小于 所述最大线速率与为所述列预设的带宽预留量之差的列的交汇点处的元素 的值进行增加, 在增加后, 所述元素所在行中各元素的和不超过所述最大 线速率, 所述元素所在列中下三角部分中的各元素的和不超过所述最大线 速率与为所述列预设的带宽预留量之差。
作为一个实施方式, 所述时隙配置模块 151 还配置为在进行时隙分配 时, 根据 OBTN环中的节点排列顺序, 从本节点开始将各节点依次作为源 节点按序进行时隙配置; 还配置为确定每个源节点下各目的节点的配置顺 序为: 根据 OBTN环中的节点排列顺序, 按照各目的节点距离所述源节点 的跳数由小到大的顺序依次进行时隙配置序。
作为一个实施方式, 所述时隙配置模块 151 还配置为在为源节点到某
一目的节点的连接分配时隙时,若在存在尚未下路的 0B的各时隙中, 同时 存在 2个以上的与当前配置的目的节点跳数最小的目的节点所占用的时隙, 则优先选用其中时隙序号最小的时隙。
实际应用中, 本发明实施例记载的上述模块和单元均可由动态带宽调 度中的中央处理器( CPU, Central Processing Unit )、数字信号处理器( DSP, Digital Signal Processor )或现场可编程门阵列 ( FPGA, Field Programmable Gate Array ) 实现。
本发明实施例还记载一种计算机存储介质, 所述计算机存储介质中存 储有计算机可执行指令, 所述计算机可执行指令用于执行以上所述的动态 带宽调度方法。
本领域普通技术人员可以理解: 实现上述方法实施例的全部或部分步 骤可以通过程序指令相关的硬件来完成, 前述的程序可以存储于一计算机 可读取存储介质中, 所述程序在执行时, 执行包括上述方法实施例的步骤; 而前述的存储介质包括: 移动存储设备、 随机存取存储器(RAM, Random Access Memory )、 只读存储器(ROM, Read-Only Memory ),磁碟或者光盘 等各种可以存储程序代码的介质。 或者, 本发明上述集成的单元如果以软 件功能模块的形式实现并作为独立的产品销售或使用时, 也可以存储在一 个计算机可读取存储介质中。 基于这样的理解, 本发明实施例的技术方案 本质上或者说对相关技术做出贡献的部分可以以软件产品的形式体现出 来, 所述计算机软件产品存储在一个存储介质中, 包括若干指令用以使得 一台计算机设备(可以是个人计算机、 服务器、 或者网络设备等)执行本 发明各个实施例所述方法的全部或部分。 而前述的存储介质包括: 移动存 储设备、 RAM、 ROM, 磁碟或者光盘等各种可以存储程序代码的介质。
以上所述, 仅为本发明的具体实施方式, 但本发明的保护范围并不局 限于此, 任何熟悉本技术领域的技术人员在本发明揭露的技术范围内, 可
轻易想到变化或替换, 都应涵盖在本发明的保护范围之内。 因此, 本发明 的保护范围应以所述权利要求的保护范围为准。
Claims
1、 一种动态带宽调度方法, 应用于光突发传送环网 0BTN中的节点, 包括:
在目标节点作为主节点时, 对于 OBTN中每一个源节点, 所述目标节 点在为所述源节点到目的节点的连接分配时隙时, 在存在尚未下路的突发 时隙 OB的各时隙中,优先选用与当前配置的目的节点跳数最小的目的节点 所占用的时隙;
在所述目标节点作为主节点时, 将时隙分配结果转化为带宽地图的形 式发送给所述 OBTN中的各从节点。
2、 如权利要求 1所述的方法, 其中, 所述方法还包括:
在所述目标节点作为主节点时, 根据全网节点的带宽请求, 为各源节 点到目的节点间的连接进行带宽授权。
3、 如权利要求 2所述的方法, 其中, 所述根据全网节点的带宽请求, 为各源节点到目的节点间的连接进行带宽授权, 包括:
所述主节点根据全网节点的带宽请求, 得到带宽请求矩阵; 其中, 所 述带宽请求矩阵中的元素 表示所述 OBTN中节点 i对节点 j的带宽请求, i和 j分别为小于等于 N的正整数, N为所述 OBTN中的节点总数;
所述主节点根据已有带宽资源对所述带宽请求矩阵进行裁减授权, 得 到带宽授权矩阵;
其中, 所述带宽授权矩阵中的元素 表示所述 OBTN中节点 i对节点 j的授权带宽, i和 j分别为小于等于 N的正整数, N为所述 OBTN中的节 点总数; 所述带宽授权矩阵中, 每一行元素的授权带宽总和小于等于最大 线速率; 对于所述带宽授权矩阵中的每一列, 所述列的上三角部分中的所 有元素的授权带宽总和, 与在一个环长周期前接收到的带宽授权矩阵中同 一列的下三角部分中的所有元素的授权带宽总和, 小于等于所述最大线速
率。
4、 如权利要求 3所述的方法, 其中, 所述主节点根据已有带宽资源对 所述带宽请求矩阵进行裁减授权, 包括:
所述主节点对所述带宽请求矩阵进行行裁减; 所述主节点对经过行裁 减后得到的矩阵进行列裁减。
5、 如权利要求 4所述的方法, 其中, 所述主节点对所述带宽请求矩阵 进行行裁减, 包括: 对于所述带宽请求矩阵中的每一行, 判断所述行中各 元素的和是否超过所述最大线速率; 若超过, 则重新对所述行中的各元素 进行带宽分配, 直至得到的所述行中的各元素的和小于等于所述最大线速 率;
所述主节点对经过行裁减后得到的矩阵进行列裁减, 包括: 对于所述 经过行裁减后得到的矩阵中的每一列, 一方面, 判断所述列的下三角部分 中的各元素的和是否超过最大线速率与为所述列预设的带宽预留量之差; 若超过, 则重新对所述列的下三角部分中的各元素进行带宽分配, 直至得 到的所述列的下三角部分中各元素的和, 小于等于所述最大线速率与为所 述列预设的带宽预留量之差; 另一方面, 判断所述列的上三角部分中的各 元素的和是否超过所述列的上三角部分的最大带宽授权量; 若超过, 则重 新对所述列的上三角部分中的各元素进行带宽分配, 直至得到的所述列中 上三角部分中各元素的和, 小于等于所述列的上三角部分的最大带宽授权 量; 其中, 所述列的上三角部分的最大带宽授权量为所述最大线速率, 与 在一个环长周期前接收到的带宽授权矩阵中的相同列的下三角部分的实际 带宽授权量之差。
6、 如权利要求 4所述的方法, 其中, 所述主节点对所述带宽请求矩 阵进行列裁减, 包括: 对于所述带宽请求矩阵中的每一列, 一方面, 判断 所述列的下三角部分中的各元素的和是否超过最大线速率与为所述列预设
的带宽预留量之差; 若超过, 则重新对所述列中的各元素进行带宽分配, 直至得到的所述列的下三角部分中各元素的和, 小于等于所述最大线速率 与为所述列预设的带宽预留量之差; 另一方面, 判断所述列的上三角部分 中的各元素的和是否超过所述列的上三角部分的最大带宽授权量; 若超过, 则重新对所述列的上三角部分中的各元素进行带宽分配, 直至得到的所述 列的上三角部分中各元素的和, 小于等于所述列的上三角部分的最大带宽 授权量; 其中, 所述列的上三角部分的最大带宽授权量, 为所述最大线速 率与在一个环长周期前接收到的带宽授权矩阵中的相同列的下三角部分的 实际带宽授权量之差;
所述主节点对经过列裁减后得到的矩阵进行行裁减, 包括: 对于所述 经过列裁减后得到的矩阵中的每一行, 判断所述行中各元素的和是否超过 所述最大线速率; 若超过, 则重新对所述行中的各元素进行带宽分配, 直 至得到的所述行中的各元素的和小于等于所述最大线速率。
7、 如权利要求 4所述的方法, 其中, 所述主节点对所述带宽请求矩阵 进行行裁减, 包括: 对于所述带宽请求矩阵中的每一行, 判断所述行中各 元素的和是否超过所述最大线速率; 若超过, 则重新对所述行中的各元素 进行带宽分配, 直至得到的所述行中的各元素的和小于等于所述最大线速 率;
所述主节点对所述带宽请求矩阵进行列裁减, 包括:
对于所述带宽请求矩阵中的每一列, 一方面, 判断所述列的下三角部 分中的各元素的和是否超过最大线速率与为所述列预设的带宽预留量之 差; 若超过, 则重新对所述列的下三角部分中的各元素进行带宽分配, 直 至得到的所述列的下三角部分中各元素的和, 小于等于所述最大线速率与 为所述列预设的带宽预留量之差; 另一方面, 判断所述列的上三角部分中 的各元素的和是否超过所述列的上三角部分的最大带宽授权量; 若超过,
则重新对所述列的上三角部分中的各元素进行带宽分配, 直至得到的所述 列中上三角部分中各元素的和, 小于等于所述列的上三角部分的最大带宽 授权量; 其中, 所述列的上三角部分的最大带宽授权量, 为所述最大线速 率与在一个环长周期前接收到的带宽授权矩阵中的相同列的下三角部分的 实际带宽授权量之差;
在确定所述带宽授权矩阵时, 取经过行裁减后得到的矩阵与经过列裁 减后得到的矩阵中位置相同的元素的值中较小的一个, 作为所述带宽授权 矩阵中相应位置的元素的值。
8、 如权利要求 7所述的方法, 其中, 所述方法还包括:
在确定了所述带宽授权矩阵中每一个元素的取值之后, 所述主节点逐 行及逐列对所述带宽授权矩阵进行扫描, 对于行中各元素的和小于所述最 大线速率的行、 及列中上三角部分中的各元素的和小于所述列的上三角部 分的最大带宽授权量的列的交汇点处的元素的值进行增加, 在增加后, 所 述元素所在行中各元素的和不超过所述最大线速率, 所述元素所在列的上 三角部分中的各元素的和不超过所述列的上三角部分的最大带宽授权量; 对于行、 及列的交汇点处的元素的值进行增加, 所述行为所包括各元素的 和小于所述最大线速率的行, 所述列满足: 所述带宽授权矩阵中下三角部 分中的各元素的和, 小于所述最大线速率与为所述列预设的带宽预留量之 差, 在增加后, 所述元素所在行中各元素的和不超过所述最大线速率, 所 述元素所在列中下三角部分中的各元素的和不超过所述最大线速率与为所 述列预设的带宽预留量之差。
9、 如权利要求 5、 6或 7所述的方法, 其中, 所述重新对所述列中的 各元素进行带宽分配, 包括:
按照固定带宽分配方式的优先级最高、 保证带宽分配方式的优先级次 之、 及非保证带宽分配方式的优先级最低的次序, 釆用当前可用的优先级
最高的带宽分配方式相应地对所述列中的各元素进行带宽分配; 所述重新对所述行中的各元素进行带宽分配, 包括:
按照固定带宽分配方式的优先级最高、 保证带宽分配方式的优先级次 之、 及非保证带宽分配方式的优先级最低的次序, 釆用当前可用的优先级 最高的带宽分配方式相应地对所述行中的各元素进行带宽分配。
10、 如权利要求 9所述的方法, 其中, 所述方法还包括:
在釆用非保证带宽分配方式对所述列或所述行中的元素进行带宽分配 后, 若所述列或行中仍有剩余带宽, 则对应将剩余带宽轮询分配至所述列 或行上未饱和的连接, 每次轮询分配 1个 OB; 其中, 未饱和的连接为既有 带宽授权量小于实际带宽请求量的连接。
11、 如权利要求 1所述的方法, 其中, 所述方法还包括:
在进行时隙分配时, 所述主节点根据 OBTN环中的节点排列顺序, 从 所述主节点开始将各节点依次作为源节点按序进行时隙配置;
每个所述源节点下各目的节点的配置顺序为: 根据 OBTN环中的节点 排列顺序, 按照各目的节点距离所述源节点的跳数由小到大的顺序依次进 行时隙配置。
12、 如权利要求 1或 11所述的方法, 其中, 所述方法还包括: 在为所述源节点到某一目的节点的连接分配时隙时, 若在存在尚未下 路的 OB的各时隙中,同时存在 2个以上的与当前配置的目的节点跳数最小 的目的节点所占用的时隙, 则优先选用时隙序号最小的时隙。
13、 如权利要求 1所述的方法, 其中, 所述方法还包括:
在所述目标节点作为从节点时, 所述目标节点对本次接收到的带宽地 图与跨环带宽地图及进行合成得到接收地图, 并根据所述接收地图对相应 数据帧进行接收; 其中, 所述跨环带宽地图为所述目标节点在一个环长周 期前接收到的带宽地图; 在所述接收地图中, 元素 RnnH 表示目的节点 n
在时隙 m接收源节点 j发出的 OB; 在素数带宽地图中, 元素 AnnH表示源 节点 n在时隙 m发送 OB至目的节点 i; m、 n、 i及 j均为正整数; 其中, 所述对本次接收到的带宽地图与跨环带宽地图及进行合成得到接收地 图, 包括:
逐行或逐列扫描所述跨环带宽地图, 针对所述跨环带宽地图中的每一 个元素, 若所述元素的取值小于所述元素所在行号, 则将所述元素的行号 作为所述接收带宽地图中以所述元素的取值及列号对应作为行号和列号的 元素的取值;
逐行或逐列扫描本次接收到的带宽地图, 针对所述带宽地图中的每一 个元素, 若所述元素的取值大于所述元素所在行号, 则将所述元素的行号 作为接收带宽地图中以所述元素的取值及列号对应作为行号和列号的元素 的取值。
14、 如权利要求 1所述的方法, 其中, 所述方法还包括:
当控制帧和对应的数据帧同时到达所述 OBTN中的节点时, 所述节点 通过光纤延时线对接收到的所述数据帧进行延迟; 其中, 延迟时间大于等 于本节点对所述控制帧的处理时间。
15、 一种光突发传送环网的动态带宽调度装置, 应用于光突发传送环 网 OBTN中的主节点上, 包括:
时隙配置模块, 配置为对于 OBTN中每一个源节点, 在为所述源节点 到目的节点的连接分配时隙时, 在存在尚未下路的突发时隙 OB 的各时隙 中, 优先选用与当前配置的目的节点跳数最小的目的节点所占用的时隙; 带宽地图转化模块, 配置为将所述时隙配置模块得到的时隙分配结果 转化为带宽地图的形式发送给所述 OBTN中各从节点。
16、 如权利要求 15所述的装置, 其中, 还包括:
带宽授权矩阵生成模块, 配置为根据全网节点的带宽请求, 得到带宽
请求矩阵; 其中, 所述带宽请求矩阵中的元素 表示所述 OBTN中节点 i 对节点 j的带宽请求, i和 j分别为小于等于 N的正整数, N为所述 OBTN 中的节点总数;
裁减授权模块, 配置为根据已有带宽资源对所述带宽请求矩阵进行裁 减授权, 得到带宽授权矩阵; 其中, 所述带宽授权矩阵中的元素 表示所 述 OBTN中节点 i对节点 j的授权带宽, i和 j分别为小于等于 N的正整数, N为所述 OBTN中的节点总数; 所述带宽授权矩阵中, 每一行元素的授权 带宽总和小于等于最大线速率; 对于所述带宽授权矩阵中的每一列, 所述 列的上三角部分中的所有元素的授权带宽总和与在一个环长周期前接收到 的带宽授权矩阵中同一列的下三角部分中的所有元素的授权带宽总和, 小 于等于所述最大线速率。
17、 如权利要求 16所述的装置, 其中, 所述裁减授权模块包括: 行裁减单元, 配置为对所述带宽请求矩阵进行行裁减;
列裁减单元, 配置为对经过行裁减后得到的矩阵进行列裁减。
18、 如权利要求 17所述的装置, 其中,
所述行裁减单元包括: 第一判断子单元, 配置为对于所述带宽请求矩 阵中的每一行, 判断所述行中各元素的和是否超过所述最大线速率; 第一 分配子单元; 配置为当所述第一判断子单元的判断结果为超过时, 重新对 所述行中的各元素进行带宽分配, 直至得到的所述行中的各元素的和小于 等于所述最大线速率;
所述列裁减单元包括: 第二判断子单元, 配置为对于所述经过行裁减 后得到的矩阵中的每一列, 判断所述列的下三角部分中的各元素的和是否 超过最大线速率与为所述列预设的带宽预留量之差; 第二分配子单元, 配 置为当所述第二判断子单元的判断结果为超过时, 重新对所述列的下三角 部分中的各元素进行带宽分配, 直至得到的所述列的下三角部分中各元素
的和小于等于所述最大线速率与为所述列预设的带宽预留量之差; 第三判断子单元, 配置为对于所述经过行裁减后得到的矩阵中的每一 列, 判断所述列的上三角部分中的各元素的和是否超过所述列的上三角部 分的最大带宽授权量; 第三分配子单元, 配置为当所述第三判断子单元的 判断结果为超过时, 则重新对所述列的上三角部分中的各元素进行带宽分 配, 直至得到的所述列中上三角部分中各元素的和小于等于所述列的上三 角部分的最大带宽授权量;
其中, 所述列的上三角部分的最大带宽授权量为所述最大线速率与在 一个环长周期前接收到的带宽授权矩阵中的相同列的下三角部分的实际带 宽授权量之差。
19、 如权利要求 17所述的装置, 其中, 所述行裁减单元包括: 第四判 断子单元, 配置为对于所述经过列裁减后得到的矩阵中的每一行, 判断所 述行中各元素的和是否超过所述最大线速率; 第四分配子单元, 配置为当 所述第四判断子单元的判断结果为超过时, 重新对所述行中的各元素进行 带宽分配, 直至得到的所述行中的各元素的和小于等于所述最大线速率; 其中,
所述列裁减单元包括: 第五判断子单元, 配置为对于所述带宽请求矩 阵中的每一列, 判断所述列的下三角部分中的各元素的和是否超过最大线 速率与为所述列预设的带宽预留量之差; 第五分配子单元, 配置为当所述 第三判断子单元的判断结果为超过时, 重新对所述列中的各元素进行带宽 分配, 直至得到的所述列的下三角部分中各元素的和小于等于所述最大线 速率与为所述列预设的带宽预留量之差;
第六判断子单元, 配置为对于所述带宽请求矩阵中的每一列, 判断所 述列的上三角部分中的各元素的和是否超过所述列的上三角部分的最大带 宽授权量; 第六分配子单元, 配置为当所述第五判断子单元的判断结果为
超过时, 重新对所述列的上三角部分中的各元素进行带宽分配, 直至得到 的所述列的上三角部分中各元素的和小于等于所述列的上三角部分的最大 带宽授权量; 其中, 所述列的上三角部分的最大带宽授权量为所述最大线 速率与在一个环长周期前接收到的带宽授权矩阵中的相同列的下三角部分 的实际带宽授权量之差。
20、 如权利要求 16所述的装置, 其中, 所述裁减授权模块还包括: 授权合成单元, 配置为在确定所述带宽授权矩阵时, 取经过行裁减后 得到的矩阵与经过列裁减后得到的矩阵中位置相同的元素的值中较小的一 个, 作为所述带宽授权矩阵中相应位置的元素的值;
所述行裁减单元包括: 第七判断子单元, 配置为对于所述带宽请求矩 阵中的每一行, 判断所述行中各元素的和是否超过所述最大线速率; 第七 分配子单元, 配置为当所述第七判断子单元的判断结果为超过时, 重新对 所述行中的各元素进行带宽分配, 直至得到的所述行中的各元素的和小于 等于所述最大线速率;
所述列裁减单元包括: 第八判断子单元, 配置为对于所述带宽请求矩 阵中的每一列, 判断所述列的下三角部分中的各元素的和是否超过最大线 速率与为所述列预设的带宽预留量之差; 第八分配子单元, 配置为当所述 第八判断子单元的判断结果为超过时, 重新对所述列的下三角部分中的各 元素进行带宽分配, 直至得到的所述列的下三角部分中各元素的和小于等 于所述最大线速率与为所述列预设的带宽预留量之差;
第九判断子单元, 配置为对于所述带宽请求矩阵中的每一列, 判断所 述列的上三角部分中的各元素的和是否超过所述列的上三角部分的最大带 宽授权量; 第九分配子单元, 配置为当所述第九判断子单元的判断结果为 超过时, 重新对所述列的上三角部分中的各元素进行带宽分配, 直至得到 的所述列中上三角部分中各元素的和小于等于所述列的上三角部分的最大
带宽授权量; 其中, 所述列的上三角部分的最大带宽授权量为所述最大线 速率与在一个环长周期前接收到的带宽授权矩阵中的相同列的下三角部分 的实际带宽授权量之差。
21、 如权利要求 20所述的装置, 其中, 所述授权合成单元包括: 扫描子单元 配置为在确定了所述带宽授权矩阵中每一个元素的取值 之后, 逐行及逐列对所述带宽授权矩阵进行扫描;
第一增加子单元, 配置为对于行中各元素的和小于所述最大线速率的 行及列中上三角部分中的各元素的和小于所述列的上三角部分的最大带宽 授权量的列的交汇点处的元素的值进行增加, 在增加后, 所述元素所在行 中各元素的和不超过所述最大线速率, 所述元素所在列的上三角部分中的 各元素的和不超过所述列的上三角部分的最大带宽授权量;
第二增加子单元, 配置为对于行中各元素的和小于所述最大线速率的 行及列中下三角部分中的各元素的和小于所述最大线速率与为所述列预设 的带宽预留量之差的列的交汇点处的元素的值进行增加, 在增加后, 所述 元素所在行中各元素的和不超过所述最大线速率, 所述元素所在列中下三 角部分中的各元素的和不超过所述最大线速率与为所述列预设的带宽预留 量之差。
22、 如权利要求 15所述的装置, 其中, 所述时隙配置模块还配置为在 进行时隙分配时, 根据 OBTN环中的节点排列顺序, 从本节点开始将各节 点依次作为源节点按序进行时隙配置; 还配置为确定每个源节点下各目的 节点的配置顺序为: 根据 OBTN环中的节点排列顺序, 按照各目的节点距 离所述源节点的跳数由小到大的顺序依次进行时隙配置序。
23、 如权利要求 15或 22所述的装置, 其中, 所述时隙配置模块还配 置为在为源节点到某一目的节点的连接分配时隙时, 若在存在尚未下路的 OB的各时隙中, 同时存在 2个以上的与当前配置的目的节点跳数最小的目
的节点所占用的时隙, 则优先选用其中时隙序号最小的时隙。
24、 一种计算机存储介质, 所述计算机存储介质中存储有计算机可执 行指令, 所述计算机可执行指令用于执行权利要求 1至 14任一项所述的动 态带宽调度方法。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP14784881.6A EP3016304B1 (en) | 2013-07-23 | 2014-06-17 | Dynamic bandwidth scheduling method and device, and computer storage medium |
US14/907,077 US9755980B2 (en) | 2013-07-23 | 2014-06-17 | Dynamic bandwidth scheduling method and device, and computer storage medium |
HK16106131.9A HK1218353A1 (zh) | 2013-07-23 | 2016-05-30 | 動態帶寬調度方法、裝置及計算機存儲介質 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310310875.7 | 2013-07-23 | ||
CN201310310875.7A CN104348551B (zh) | 2013-07-23 | 2013-07-23 | 一种光突发传送环网的动态带宽调度方法和装置 |
Publications (2)
Publication Number | Publication Date |
---|---|
WO2014169884A2 true WO2014169884A2 (zh) | 2014-10-23 |
WO2014169884A3 WO2014169884A3 (zh) | 2014-12-11 |
Family
ID=51731908
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2014/080027 WO2014169884A2 (zh) | 2013-07-23 | 2014-06-17 | 动态带宽调度方法、装置及计算机存储介质 |
Country Status (5)
Country | Link |
---|---|
US (1) | US9755980B2 (zh) |
EP (1) | EP3016304B1 (zh) |
CN (1) | CN104348551B (zh) |
HK (1) | HK1218353A1 (zh) |
WO (1) | WO2014169884A2 (zh) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114826925A (zh) * | 2022-04-01 | 2022-07-29 | 西北大学 | 一种数据中心间基于成本时序的大数据传输带宽调度方法 |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105744385B (zh) * | 2014-12-10 | 2020-06-16 | 中兴通讯股份有限公司 | 一种光突发传送网的传输方法和*** |
CN106209686B (zh) * | 2015-04-30 | 2019-11-05 | 南京中兴软件有限责任公司 | 一种光突发环网动态带宽分配方法及装置 |
JP6418194B2 (ja) * | 2016-03-30 | 2018-11-07 | トヨタ自動車株式会社 | 無線通信装置および無線通信方法 |
CN106685521B (zh) * | 2017-01-13 | 2020-01-10 | 国家电网公司 | 一种用于光通信网的资源余量预警的方法及装置 |
CN111277316B (zh) * | 2019-12-27 | 2021-05-25 | 北京邮电大学 | 一种基于光突发交换网络的数据传输方法、装置及*** |
CN115696466B (zh) * | 2022-09-23 | 2023-06-06 | 广州爱浦路网络技术有限公司 | 一种s-nssai的优先级确定方法及装置 |
Family Cites Families (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6259704B1 (en) * | 1996-04-23 | 2001-07-10 | Hitachi Denshi Kabushiki Kaisha | Digital transmission network |
US7009991B2 (en) * | 2002-03-28 | 2006-03-07 | Matisse Networks | Reservation-based media access controller and reservation-based optical network |
US7286759B1 (en) * | 2002-04-12 | 2007-10-23 | Alcatel Lucent | System and method for dynamic wavelength assignment in wavelength division multiplex ring networks |
KR100474694B1 (ko) * | 2002-10-12 | 2005-03-10 | 삼성전자주식회사 | 버스트 데이터 통신을 위한 링형 광네트워크 |
KR100459572B1 (ko) * | 2002-10-15 | 2004-12-03 | 삼성전자주식회사 | 버스트 통신을 위한 다중 링형 광 네트워크 |
US7345998B2 (en) * | 2004-12-15 | 2008-03-18 | Smart Labs, Inc. | Mesh network of intelligent devices communicating via powerline and radio frequency |
US7630363B2 (en) | 2004-12-31 | 2009-12-08 | Samsung Electronics Co., Ltd. | Apparatus for adjusting receiving time point of burst data in optical burst switching network and method thereof |
CN100546227C (zh) * | 2005-12-02 | 2009-09-30 | 真宽通信技术(苏州)有限公司 | 以太网无源光网络多业务动态带宽分配的方法 |
CN101193056B (zh) * | 2006-11-22 | 2010-05-12 | 中兴通讯股份有限公司 | 光突发交换设备中解决突发冲突的方法 |
CN101212403B (zh) * | 2006-12-27 | 2010-11-10 | 华为技术有限公司 | 实现光突发交换环网的方法和节点 |
CN101557339B (zh) * | 2008-04-09 | 2012-04-04 | 华为技术有限公司 | 一种数据传输方法、数据处理节点以及数据传输*** |
CN101883314B (zh) * | 2009-05-08 | 2013-05-22 | 富士通株式会社 | 用于无线通信***的媒介接入方法和设备 |
CN101895367A (zh) * | 2009-05-22 | 2010-11-24 | 华为技术有限公司 | 一种数据传输方法、网络设备和*** |
CN101959083B (zh) * | 2009-07-21 | 2013-12-04 | 华为技术有限公司 | 数据处理方法和数据处理设备 |
CN102118212B (zh) * | 2009-12-30 | 2014-04-09 | 中兴通讯股份有限公司 | 一种进行光网络单元带宽分配的方法及光线路终端 |
EP2375773A1 (en) * | 2010-03-29 | 2011-10-12 | Intune Networks Limited | Random GAP insertion in an optical ring network |
US9094126B2 (en) * | 2012-11-07 | 2015-07-28 | Nec Laboratories America, Inc. | QoS-aware united control protocol for optical burst switching in software defined optical networks |
-
2013
- 2013-07-23 CN CN201310310875.7A patent/CN104348551B/zh active Active
-
2014
- 2014-06-17 EP EP14784881.6A patent/EP3016304B1/en active Active
- 2014-06-17 US US14/907,077 patent/US9755980B2/en active Active
- 2014-06-17 WO PCT/CN2014/080027 patent/WO2014169884A2/zh active Application Filing
-
2016
- 2016-05-30 HK HK16106131.9A patent/HK1218353A1/zh unknown
Non-Patent Citations (1)
Title |
---|
None |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114826925A (zh) * | 2022-04-01 | 2022-07-29 | 西北大学 | 一种数据中心间基于成本时序的大数据传输带宽调度方法 |
CN114826925B (zh) * | 2022-04-01 | 2023-05-09 | 西北大学 | 一种数据中心间基于成本时序的大数据传输带宽调度方法 |
Also Published As
Publication number | Publication date |
---|---|
CN104348551B (zh) | 2018-12-28 |
CN104348551A (zh) | 2015-02-11 |
US9755980B2 (en) | 2017-09-05 |
EP3016304A2 (en) | 2016-05-04 |
EP3016304A4 (en) | 2016-07-13 |
WO2014169884A3 (zh) | 2014-12-11 |
EP3016304B1 (en) | 2017-08-16 |
HK1218353A1 (zh) | 2017-02-10 |
US20160366064A1 (en) | 2016-12-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2014169884A2 (zh) | 动态带宽调度方法、装置及计算机存储介质 | |
US7965732B2 (en) | Scheduling token-controlled data transmissions in communication networks | |
US7606262B1 (en) | Universal edge node | |
JP4853821B2 (ja) | 局側装置における帯域割当装置、帯域割当方法および帯域割当プログラム | |
CN102957689B (zh) | 双层网络中的调度 | |
Kaheel et al. | Quality-of-service mechanisms in IP-over-WDM networks | |
US20040156367A1 (en) | Hierarchically distributed scheduling apparatus and method | |
US20050207427A1 (en) | Token-controlled data transmissions in communication networks | |
CN104348754B (zh) | 一种光突发环网的带宽分配方法和装置 | |
AU2010215830A1 (en) | Flexible reservation request and scheduling mechanisms in a managed shared network with quality of service | |
US8064341B2 (en) | Temporal-spatial burst switching | |
US7529267B2 (en) | Data transmissions in communication networks using multiple tokens | |
WO2012109847A1 (zh) | 一种通信网络数据传输方法、节点和*** | |
CN109150722A (zh) | 一种业务转发的方法及网络设备 | |
KR100503417B1 (ko) | 이더넷 수동형 광 네트워크에서의 QoS 보장형 스케쥴링시스템 및 방법 | |
WO2014187425A1 (zh) | 跨主节点业务处理方法和装置 | |
WO2019080744A1 (zh) | 控制通道实现方法、装置、设备、存储介质和处理方法 | |
KR100986224B1 (ko) | 이더넷 수동광가입자망에서의 동적 대역할당 장치 및 그 방법 | |
CN115550161A (zh) | 参数配置的方法、装置、控制器、通信设备和通信*** | |
Radivojević et al. | WDM EPON Architecture | |
Alexander | A protocol for high speed optical networks from a basis of satellite protocol designs | |
Kam et al. | Real-time distributed scheduling algorithm for supporting QoS over WDM networks | |
Mollard et al. | COSC460 Report: Medium access control protocols for WDM optical networks | |
Asghari et al. | Author's Accepted Manuscript | |
Hu | An Arrangement of Channels and Transceivers in Optical Packet Switching Networks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 14784881 Country of ref document: EP Kind code of ref document: A2 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14907077 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
REEP | Request for entry into the european phase |
Ref document number: 2014784881 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2014784881 Country of ref document: EP |