CN112350949A - Rerouting congestion control method and system based on flow scheduling in software defined network - Google Patents
Rerouting congestion control method and system based on flow scheduling in software defined network Download PDFInfo
- Publication number
- CN112350949A CN112350949A CN202011149218.5A CN202011149218A CN112350949A CN 112350949 A CN112350949 A CN 112350949A CN 202011149218 A CN202011149218 A CN 202011149218A CN 112350949 A CN112350949 A CN 112350949A
- Authority
- CN
- China
- Prior art keywords
- link
- flow
- path
- congestion
- scheduling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
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
- H04L47/11—Identifying congestion
-
- 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
- H04L47/12—Avoiding congestion; Recovering from congestion
-
- 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
- H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
- H04L47/2425—Traffic characterised by specific attributes, e.g. priority or QoS for supporting services specification, e.g. SLA
- H04L47/2433—Allocation of priorities to traffic types
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
- H04L47/52—Queue scheduling by attributing bandwidth to queues
- H04L47/525—Queue scheduling by attributing bandwidth to queues by redistribution of residual bandwidth
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate
Abstract
The invention relates to the technical field of communication, in particular to a rerouting congestion control method and a rerouting congestion control system based on flow scheduling in a software defined network, which comprises the following steps: acquiring network state and switch port flow information, and transmitting the acquired data to a data processing module; calculating the link utilization rate according to the acquired data, judging whether the link is congested or not according to the link utilization rate, and if the link is congested, selecting the flow on the congested link to perform rerouting scheduling to relieve the congestion; if no congestion occurs, no processing is performed. The invention adopts the CC-KP algorithm to simplify the congestion relieving algorithm of flow scheduling after the occurrence of congestion into two steps of scheduling large flow selection and rerouting path selection. By rerouting and scheduling the large flow on the congested link, the network flow transmission performance can be improved and the large flow completion time can be reduced while network congestion is effectively relieved.
Description
Technical Field
The invention relates to the technical field of communication, in particular to a rerouting congestion control method and a rerouting congestion control system based on flow scheduling in a software defined network.
Background
With the explosive growth of big data and cloud computing services, large data centers are rapidly developing. The traffic of the data center is increased explosively, the internal network topology is more and more complex, the scale is larger and larger, and the traditional network congestion management gradually exposes some insufficient places. The emergence of Software Defined Networking (SDN) provides a new idea for load imbalance, link redundancy, and other problems. Compared with the traditional network, the separation of the SDN control plane and the forwarding plane is beneficial to solving the congestion management problem of the network due to the characteristic of decoupling control and forwarding. Therefore, how to manage network congestion based on an SDN architecture, process complex and huge traffic of a scheduling data center, and how to fully utilize network link resources becomes a hot issue of network research.
Data center traffic is mostly internal east-west traffic, which provides convenience for centralized management and is also beneficial to function expansion of a control plane. The congestion research directions based on flow scheduling in the prior art are roughly divided into two types: a backup path routing method and a rerouting scheduling method. Compared with a static standby path routing method, the congestion can not be dynamically scheduled according to the network condition in real time, and the congestion is managed by the dynamic rerouting flow scheduling by better utilizing the centralized control characteristic of the SDN. The application of the multi-path topology such as Fatree, BCube, VL2 and the like in the data center network also provides multiple choices for the selection of the rerouting path.
At present, common congestion management schemes include a congestion solution based on a flow scheduling cost, a cubic c-PEF algorithm based on link criticality, a congestion solution based on an equivalent multi-path (ECMP) algorithm, and the like, wherein the congestion solution based on the flow scheduling cost performs rerouting scheduling on a flow with a minimum scheduling cost when a network is congested, and the algorithm can effectively relieve link congestion. However, when the network becomes large, the complexity of the algorithm increases, and the network links become too congested, the efficiency of congestion relief may suffer. The main idea of the link criticality-based square CC-PEF algorithm is that a controller periodically sends status polling messages, when a link is loaded, large flows on a key link are scheduled, and network congestion is specifically solved in a flow scheduling mode. However, the implementation of the algorithm increases the overhead of the controller and the switch, and the careful flow scheduling algorithm also affects the load handling rate of the congested link. An Equal Cost Multi Path (ECMP) algorithm uses static hashing to distribute a flow to multiple equal cost paths for transmission, but when the bandwidth of the paths is too different, the efficiency is reduced. In summary, the main problems faced by the current congestion control algorithm include three types: how to efficiently alleviate congestion; link resources are utilized as much as possible to improve the network flow transmission performance of the data center; the load of the controller and the switch is reduced as much as possible.
Disclosure of Invention
In order to solve the above problems, the present invention provides a rerouting congestion control method and system based on flow scheduling in a software defined network.
A rerouting congestion control method based on flow scheduling in a software defined network comprises the following steps:
s1, acquiring network state and switch port flow information, and transmitting the acquired data to a data processing module;
s2, calculating the link utilization ratio according to the obtained data, judging whether the link is congested or not according to the link utilization ratio, and if so, selecting the flow on the congested link to carry out rerouting scheduling to relieve congestion; if the congestion does not occur, the processing is not carried out;
the method for selecting the flow on the congestion link to carry out rerouting scheduling to relieve the congestion comprises the following steps:
s21, adopting CC-KP algorithm to screen out the congestion link emnLarge set of flows on F ═ F1,f2,......,fn}；
S22, calculating the dispatching priority of each flow
Selecting from a large stream set
The largest flow is used as the large flow f to be scheduledk；
S23、Solving a short path set P by adopting a Yen-K algorithmk；
S24, calculating the residual bandwidth of each path, if the path residual bandwidth
Less than large flow f to be scheduledkIf so, the path is not considered; if the path has residual bandwidth
If the bandwidth requirement of the scheduled big stream is greater than or equal to the bandwidth requirement of the scheduled big stream, calculating the path selection priority G of the pathiFrom a short path set PkIn (1) is selected from GiThe largest path is taken as a rerouting path (i.e. an optimal scheduling path);
s25, the controller issues the flow table to the switch, and the large flow f to be scheduled is sent to the switchkRerouting to reroute paths to relieve congestion and increasing load on corresponding links of the reroute pathsk。
Further, the calculation formula of the link utilization rate includes:
wherein the content of the first and second substances,
indicating the utilization, load, of links between nodes in the pathmnIndicating a link e at a certain timemnLoad of (B)mnRepresents a link emnThe bandwidth size of (d).
Further, controlling the polling cycle includes: and acquiring network state and switch port flow information through OpenFlow periodic polling messages of a data link layer.
Further, judging whether the link is congested according to the link utilization ratio specifically includes: when the link utilization rate is greater than the congestion threshold of the link, judging that the link is congested; and when the link utilization rate is less than or equal to the congestion threshold value of the link, judging that the link is not congested.
A system for flow scheduling based rerouting congestion control in a software defined network, comprising: the system comprises a network state monitoring module, a scheduling flow selection module and a rerouting path selection module, wherein the network state monitoring module is mainly used for acquiring network state and switch port flow information; the scheduling flow selection module firstly screens out the large flow on the congestion link by a large flow detection method at a host end, and then selects the large flow suitable for scheduling according to the survival time of the flow; the rerouting path selection module firstly obtains a short path set through a Yen-k algorithm, and then selects a proper scheduling path from the short path set through path flatness and link residual bandwidth.
The invention has the beneficial effects that:
1. the invention provides a flow scheduling algorithm CC-KP algorithm based on a rerouting mechanism, which fully exerts the advantages of SDN control centralization and network state monitoring, combines the characteristics of data center network topology multipath, can schedule large flows on congested links, can quickly relieve link congestion under the condition of reducing the expenses of a controller and a switch as much as possible, and improves the data flow transmission performance of the whole network.
And 2, simplifying a congestion relieving algorithm of flow scheduling after congestion occurs into two steps of scheduling large flow selection and rerouting path selection by the CC-KP algorithm. By rerouting and scheduling the large flow on the congested link, the network flow transmission performance can be improved and the large flow completion time can be reduced while network congestion is effectively relieved.
Drawings
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
FIG. 1 is a flowchart of a CC-KP algorithm according to an embodiment of the present invention;
fig. 2 is a table showing the header of a packet segment according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
According to the method, under the SDN framework, according to the data center network flow characteristics and the network topology multi-path characteristics, the CC-KP algorithm is adopted to dispatch the heavy flow rerouting on the congested link to the low-load link, and under the condition that the expenses of a controller and an exchanger are reduced as much as possible, the network congestion can be effectively relieved, the network flow transmission performance is improved, and the heavy flow completion time is reduced.
As shown in fig. 1, the subject idea of the CC-KP algorithm includes: when the network is congested, the rerouting on the congested link is transmitted to a new path with a lower load, so that the congested data flow can be quickly relieved.
The CC-KP algorithm is roughly divided into three steps: monitoring the network congestion state, selecting a scheduling flow and selecting a rerouting path.
Firstly, the main monitoring content of the network congestion state is to monitor the load of a link in the network, and when the load of the link in the network exceeds a set threshold value, a CC-KP algorithm is started to relieve the congestion.
And acquiring the network link topology by utilizing a controller through an LLDP protocol based on the SDN architecture. The network state and the switch port flow information are obtained through the regular polling message of the network communication protocol OpenFlow of the data link layer, the obtained data are transmitted to the data processing module, the polling period is controlled during the data obtaining period, and the overload is avoided. The control polling cycle mainly includes: network state and switch port flow information are obtained through regular polling messages of OpenFlow of a data link layer, and load pressure of a switch and a controller can be effectively relieved.
Calculating the link utilization rate according to the acquired data, judging whether the link is congested or not according to the link utilization rate, and if the link is congested, selecting the flow on the congested link to perform rerouting scheduling to relieve the congestion; if no congestion occurs, no processing is performed.
The calculation of the link utilization includes: setting G (V, E) of the data center network, wherein V is the set of all nodes in the network, E is the set of all links in the topology, P is the set of all paths from a source node i to a destination node j,
indicating the link utilization for the ith path,
indicating the utilization of the links between the various nodes in the path. For calculating the utilization rate of each link, load is usedmnIndicating a link e at a certain timemnI.e. the size of the bandwidth occupied by the transmission data stream, BmnRepresents a link emnThe bandwidth size of (d). The link utilization calculation formula is as follows:
by ηthIndicating link congestion threshold when link utilization
Greater than link congestion threshold ηthJudging that the link is congested; and when the link utilization rate is less than or equal to the congestion threshold value of the link, judging that the link is not congested.
Further, in one embodiment, the link congestion threshold ηthThe setting of (1) comprises: when the link congestion threshold is set to be too large, the network packet loss rate is increased, and when the link congestion threshold is set to be too low, the network resources are consumed by insufficient utilization of the critical link bandwidth and frequent flow scheduling. In a preferred embodiment of the invention, the link congestion threshold eta of the CC-KP algorithm is usedthThe result was set to 90%.
If the link is judged to be congested according to the link utilization rate, selecting the flow on the congested link to carry out rerouting scheduling to relieve the congestion, and specifically comprising the following steps of:
s21, adopting CC-KP algorithm to screen out the congestion link emnLarge set of flows on F ═ F1,f2,......,fn}；
S22, calculating the dispatching priority of each flow
Selecting from a large stream set
The largest flow is used as the large flow f to be scheduledk；
S23, adopting Yen-K algorithm to obtain short path set Pk；
S24, calculating the residual bandwidth of each path, if the path residual bandwidth
Less than large flow f to be scheduledkIf so, the path is not considered; if the path has residual bandwidth
If the bandwidth requirement of the scheduled big stream is greater than or equal to the bandwidth requirement of the scheduled big stream, calculating the path selection priority G of the pathiFrom a short path set PkIn (1) is selected from GiThe largest path is taken as a rerouting path (i.e. an optimal scheduling path);
s25, the controller issues a flow table, and the large flow f to be scheduled is sentkRerouting and scheduling to reroute paths to relieve congestion and increasing load of corresponding links in the scheduled pathsk。
Secondly, the selection step of the scheduling flow solves the problem of selecting the flow which can quickly relieve congestion for the rerouting flow scheduling, and the specific implementation process comprises the following steps:
(1) detection of a large flow: studies have shown that approximately 90% of the flows in a data center network are latency sensitive small flows less than 100kb, whereas large flows, accounting for only 10% in number, occupy more than 90% of the bytes, and have a longer lifetime. Considering that the installation time of one flow table in the OpenFlow switch is about 1-10 ms and the time overhead of a scheduling algorithm, it is obvious that a large flow with a long survival time has a higher scheduling value when congestion occurs. The current method for judging the large flow mostly utilizes a controller and a switch to detect the large flow. For example, the controller obtains byte information of the flow from the switch to perform large flow detection centrally, but this increases the load of the controller and also occupies network bandwidth resources when polling information is too frequent. It is also possible to record the data statistics of each flow through the switch using the switch flow entry, but doing so would occupy valuable flow entry resources of the switch. Secondly, the bandwidth between the switch and the controller is limited, so that transmitting statistics in the data centre network becomes a new burden for traffic management. The timely delivery of statistical information and the load on the controller in these methods can affect the detection efficiency.
Based on the advantages and disadvantages of the detection scheme in the prior art, the invention adopts the CC-KP algorithm to screen out the congested link emnLarge set of flows on F ═ F1,f2,......,fn}. The CC-KP algorithm adopts a large flow detection scheme based on a host side, so that the load of a switch and a controller can be reduced while the detection rate is improved. The specific process of adopting CC-KP algorithm screening comprises the following steps: monitoring a TCP Socket cache at a host end through a shim over each terminal host, marking a stream exceeding a threshold value, and marking a subsequent data packet of the stream by using an in-band signaling mechanism (DSCP), namely setting a DSCP bit in the data packet to xxxx11 by setting a differentiated service Domain (DSCP) of a data packet header, as shown in FIG. 2, a list for representing the header of a packet segment is provided.
(2) Selecting a scheduling flow: using the set F { F1,f2,...fk...,fnDenotes a link emnA large flow of water. By using
Indicating the scheduling priority of the kth big stream by
Indicating the specific size of the kth stream by
Indicating the lifetime of the kth large stream. And meanwhile, the number of the congestion links through which a large flow flows is recorded as n, and the congestion can be relieved more quickly by scheduling the flows through a plurality of congestion links when the congestion occurs. The specific calculation formula of the scheduling priority is as follows.
To quickly relieve congestion, scheduling prioritization of large flows on congested links,
a larger flow means that the flow has a larger impact on the link congestion and is therefore scheduled preferentially.
Thirdly, the step of selecting the rerouting path is to select a proper new path for the scheduling flow by calculating the residual bandwidth and the path balance degree of the link, and the specific implementation process comprises the following steps:
(1) selecting a short path set: the Yen-K algorithm is an extension of the shortest path algorithm, and K-1 deviation short paths are deviated through the shortest path by utilizing a deviation path algorithm thought in a recurrence method. The shortest path and K-1 path sets of secondary short paths, namely a short path set P are obtained by a Yen-K algorithm with the hop number of the path route as a parameterk。
Solving a short path set P by adopting a Yen-K algorithmkThe steps and methods are not the creation points of the present invention, so the present embodiment is not described in detail, and reference can be made to the prior art documents: YEN J Y.A. algorithm for defining short routes from all sources nodes to a digit destination in general networks [ J].Quarterly of Applied Mathematics,1970,27(4):526。
(2) Selection of optimal scheduling pathTaking: it is obviously not sufficient to consider a path as a rerouting path only from the number of route hops, the present invention comprehensively considers various parameters of the link, including the link residual bandwidth and the path selection priority, and then collects the path from the path set PkAnd screening through the weight to finally obtain the rerouting path. Specifically, whether the residual bandwidth of the path meets the requirement is judged firstly, then the path selection priority G is calculated, the larger the path selection priority G is, the more suitable the path is as a rerouting path, and the short path set P is selected from the path selection priority GkIn (1) is selected from GiThe largest path is taken as the rerouted path (i.e., the optimal scheduled path).
Further, in some embodiments, the link parameters considered in selecting the optimal scheduling path in the present invention include: the method comprises the following steps of:
firstly, calculating the residual bandwidth of the link, wherein the calculation formula is as follows:
wherein the content of the first and second substances,
represents a link emnResidual bandwidth size of, BmnRepresents a link emnBandwidth size, load ofmnIndicating a link e at a certain timemnThe load of (2).
By using
Indicating the remaining bandwidth of the ith path,
if the path has residual bandwidth
Less than large flow f to be scheduledkThe bandwidth requirement of (1), the path is not considered; if the path has residual bandwidth
Is greater than or equal to the large flow f to be scheduledkThen calculating the path selection priority Gi。
Further, in some embodiments, the path selection priority GiThe calculation of (a) includes: introducing parameter bandwidth residual rate RiThe bandwidth surplus ratio calculation formula of the ith path is as follows:
where t represents the number of links included in the path.
The bandwidth difference of each link in the path is considered, the flatness of the parameter path is introduced, and A is usediExpressing the flatness of the ith path, the expression is as follows:
Ai＝MIN(Bi,1,B1,2,......,Bm,n,Bn,j)/MAX(Bi,1,B1,2,......,Bm,n,Bn,j)
according to the above path flatness AiAnd bandwidth residual rate RiCalculating a path selection priority GiThe calculation formula of the path selection priority is as follows:
Gi＝Ai+Ri
the path selection algorithm proceeds from a short path set PkThe method considers the path state to select a proper scheduling path, avoids selecting the scheduling path from all reachable paths, and optimizes the algorithm complexity.
A rerouting congestion control system based on flow scheduling in a software defined network comprises a monitoring module of a network state, a scheduling flow selection module and a rerouting path selection module.
The network state monitoring module is mainly used for acquiring network state and exchanger port flow information and providing information for next flow scheduling.
The dispatching flow selecting module firstly screens out the large flow on the congestion link by a large flow detection method at the host end, and then selects the large flow suitable for dispatching according to the survival time of the flow.
The rerouting path selection module firstly obtains a short path set through a Yen-k algorithm, and then screens out a proper scheduling path from the short path set through path flatness and link residual bandwidth.
In order to verify the effectiveness of the algorithm, a minnet network simulation environment is built under the ubuntu system, and RYU is selected as a network controller.
The RYU is a common light-weight power controller, and a congestion judgment module, a scheduling flow selection module and a rerouting module are realized in the RYU controller through programming. The flow is generated by using an Iperf tool in a simulation mode, the flow model is set to be 90% of small flow with the size of 1kb-100kb and 10% of large flow with the size of 100kb-100mb, the flow is uniformly distributed, and the arrival process of the flow belongs to a Poisson process. The Fat tree structure is characterized in that aggregation layer switches and edge layer switches can form a Pod, the topology has K pods, each Pod has K/2 aggregation layer switches and K/2 edge layer switches, the number of core layer switches is (K/2) ^2, and each core switch has K ports. We set K of the Yen-K algorithm here to 4 according to the topological link characteristics.
The effect of the evaluation algorithm is verified through three indexes of congestion relieving time, leveling average bandwidth utilization rate and large leveling average completion time in the experiment. The congestion relief time refers to the time required for a certain link to reach a congestion threshold value and generate congestion until the congestion is relieved, the bandwidth utilization rate of the flow refers to the ratio of the bandwidth of the flow at the receiving end to the bandwidth of the flow at the sending end, and because of the influence of network state fluctuation, the actual transmission rate of the flow is often smaller than the generation rate, so the transmission performance of the network link flow can be reflected through the ratio. The average completion time of a flow refers to the average time delay of the flow in the network from the edge switch at the transmitting end to the edge switch at the receiving end through each switch and each link. The estimation algorithm can effectively relieve network congestion, improve the transmission performance of the network flow and reduce the completion time of the large flow.
It should be noted that, as one of ordinary skill in the art would understand, all or part of the processes of the above method embodiments may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when executed, the computer program may include the processes of the above method embodiments. The storage medium may be a magnetic disk, an optical disk, a Read-0nly Memory (ROM), a Random Access Memory (RAM), or the like.
The foregoing is directed to embodiments of the present invention and it will be appreciated by those skilled in the art that changes may be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (5)
1. A rerouting congestion control method based on flow scheduling in a software defined network is characterized by comprising the following steps:
s1, acquiring network state and switch port flow information, and transmitting the acquired data to a data processing module;
s2, calculating the link utilization ratio according to the obtained data, judging whether the link is congested or not according to the link utilization ratio, and if so, selecting the flow on the congested link to carry out rerouting scheduling to relieve congestion; if the congestion does not occur, the processing is not carried out;
the method for selecting the flow on the congestion link to carry out rerouting scheduling to relieve the congestion comprises the following steps:
s21, adopting CC-KP algorithm to screen out the congestion link emnLarge set of flows on F ═ F1,f2,......,fn}；
S22, calculating the dispatching priority of each flow
Selecting from a large stream set
The largest flow is used as the large flow f to be scheduledk；
S23, adopting Yen-K algorithm to obtain short path set Pk；
S24, calculating the residual bandwidth of each path, if the path residual bandwidth
Less than large flow f to be scheduledkIf so, the path is not considered; if the path has residual bandwidth
If the bandwidth requirement of the scheduled big stream is greater than or equal to the bandwidth requirement of the scheduled big stream, calculating the path selection priority G of the pathiFrom a short path set PkIn (1) is selected from GiThe largest path is taken as a rerouting path (i.e. an optimal scheduling path);
s25, the controller issues the flow table to the switch, and the large flow f to be scheduled is sent to the switchkRerouting to reroute paths to relieve congestion and increasing load on corresponding links of the reroute pathsk。
2. The method of claim 1, wherein the calculation formula of link utilization comprises:
wherein the content of the first and second substances,
indicating the utilization, load, of links between nodes in the pathmnIndicating a link e at a certain timemnLoad of (B)mnRepresents a link emnThe bandwidth size of (d).
3. The method of claim 1, wherein controlling the polling period comprises: and acquiring network state and switch port flow information through OpenFlow periodic polling messages of a data link layer.
4. The method for controlling rerouting congestion in a software defined network based on flow scheduling according to claim 1, wherein judging whether a link is congested according to a link utilization ratio specifically includes: when the link utilization rate is greater than the congestion threshold of the link, judging that the link is congested; and when the link utilization rate is less than or equal to the congestion threshold value of the link, judging that the link is not congested.
5. A system for flow scheduling based rerouting congestion control in a software defined network, comprising: a monitoring module of network state, a scheduling flow selecting module and a rerouting path selecting module,
the network state monitoring module is mainly used for acquiring network state and switch port flow information;
the scheduling flow selection module firstly screens out the large flow on the congestion link by a large flow detection method at a host end, and then selects the large flow suitable for scheduling according to the survival time of the flow;
the rerouting path selection module firstly obtains a short path set through a Yen-k algorithm, and then selects a proper scheduling path from the short path set through path flatness and link residual bandwidth.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011149218.5A CN112350949B (en) | 2020-10-23 | 2020-10-23 | Rerouting congestion control method and system based on flow scheduling in software defined network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011149218.5A CN112350949B (en) | 2020-10-23 | 2020-10-23 | Rerouting congestion control method and system based on flow scheduling in software defined network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112350949A true CN112350949A (en) | 2021-02-09 |
| CN112350949B CN112350949B (en) | 2022-05-31 |
Family
ID=74360128
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011149218.5A Active CN112350949B (en) | 2020-10-23 | 2020-10-23 | Rerouting congestion control method and system based on flow scheduling in software defined network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112350949B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113132180A (en) * | 2021-03-11 | 2021-07-16 | 武汉大学 | Cooperative type large flow detection method facing programmable network |
| CN114051001A (en) * | 2021-11-10 | 2022-02-15 | 中国电信股份有限公司 | Flow data processing method and device, storage medium and electronic equipment |
| CN114827036A (en) * | 2022-04-18 | 2022-07-29 | 天津大学 | NDN hop-by-hop congestion control method with cache perception based on SDN |
| CN115150324A (en) * | 2022-06-09 | 2022-10-04 | 南京邮电大学 | Method and system for realizing variable rerouting threshold based on programmable data plane |
| CN116032829A (en) * | 2023-03-24 | 2023-04-28 | 广东省电信规划设计院有限公司 | SDN network data stream transmission control method and device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150326426A1 (en) * | 2014-05-12 | 2015-11-12 | Futurewei Technologies, Inc. | Partial software defined network switch replacement in ip networks |
| CN105227481A (en) * | 2015-09-02 | 2016-01-06 | 重庆邮电大学 | The SDN congestion control method for routing of cost minimization is dispatched based on path cost and stream |
| CN106357547A (en) * | 2016-09-08 | 2017-01-25 | 重庆邮电大学 | Software-defined network congestion control algorithm based on stream segmentation |
| CN106533960A (en) * | 2016-12-23 | 2017-03-22 | 重庆邮电大学 | Data center network routing method based on Fat-Tree structure |
| CN109547340A (en) * | 2018-12-28 | 2019-03-29 | 西安电子科技大学 | SDN data center network jamming control method based on heavy-route |
| CN109756421A (en) * | 2019-01-23 | 2019-05-14 | 华南理工大学 | A kind of congestion control system and method based on OpenFlow technology |
| US20200136972A1 (en) * | 2018-10-27 | 2020-04-30 | Cisco Technology, Inc. | Congestion notification reporting for a responsive network |
-
2020
- 2020-10-23 CN CN202011149218.5A patent/CN112350949B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150326426A1 (en) * | 2014-05-12 | 2015-11-12 | Futurewei Technologies, Inc. | Partial software defined network switch replacement in ip networks |
| CN105227481A (en) * | 2015-09-02 | 2016-01-06 | 重庆邮电大学 | The SDN congestion control method for routing of cost minimization is dispatched based on path cost and stream |
| CN106357547A (en) * | 2016-09-08 | 2017-01-25 | 重庆邮电大学 | Software-defined network congestion control algorithm based on stream segmentation |
| CN106533960A (en) * | 2016-12-23 | 2017-03-22 | 重庆邮电大学 | Data center network routing method based on Fat-Tree structure |
| US20200136972A1 (en) * | 2018-10-27 | 2020-04-30 | Cisco Technology, Inc. | Congestion notification reporting for a responsive network |
| CN109547340A (en) * | 2018-12-28 | 2019-03-29 | 西安电子科技大学 | SDN data center network jamming control method based on heavy-route |
| CN109756421A (en) * | 2019-01-23 | 2019-05-14 | 华南理工大学 | A kind of congestion control system and method based on OpenFlow technology |
Non-Patent Citations (2)
| Title |
|---|
| RENUGA KANAGEVLU等: "SDN Controlled Local Re-routing to Reduce Congestion in Cloud Data Center", 《2015 INTERNATIONAL CONFERENCE ON CLOUD COMPUTING RESEARCH AND INNOVATION (ICCCRI)》 * |
| 张宇巍: "软件定义数据中心网络中的拥塞控制机制研究", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113132180A (en) * | 2021-03-11 | 2021-07-16 | 武汉大学 | Cooperative type large flow detection method facing programmable network |
| CN113132180B (en) * | 2021-03-11 | 2022-07-29 | 武汉大学 | Cooperative type large flow detection method facing programmable network |
| CN114051001A (en) * | 2021-11-10 | 2022-02-15 | 中国电信股份有限公司 | Flow data processing method and device, storage medium and electronic equipment |
| CN114827036A (en) * | 2022-04-18 | 2022-07-29 | 天津大学 | NDN hop-by-hop congestion control method with cache perception based on SDN |
| CN114827036B (en) * | 2022-04-18 | 2023-09-29 | 天津大学 | SDN-based NDN hop-by-hop congestion control method with cache perception |
| CN115150324A (en) * | 2022-06-09 | 2022-10-04 | 南京邮电大学 | Method and system for realizing variable rerouting threshold based on programmable data plane |
| CN116032829A (en) * | 2023-03-24 | 2023-04-28 | 广东省电信规划设计院有限公司 | SDN network data stream transmission control method and device |
| CN116032829B (en) * | 2023-03-24 | 2023-07-14 | 广东省电信规划设计院有限公司 | SDN network data stream transmission control method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112350949B (en) | 2022-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112350949B (en) | Rerouting congestion control method and system based on flow scheduling in software defined network | |
| CN108512760B (en) | Routing method based on SDN guarantee service QoS | |
| Zhang et al. | Load balancing in data center networks: A survey | |
| EP2911348B1 (en) | Control device discovery in networks having separate control and forwarding devices | |
| CN107579922B (en) | Network load balancing device and method | |
| US6310881B1 (en) | Method and apparatus for network control | |
| CN108289064B (en) | Hybrid load balancing method in data center network | |
| CN105471764B (en) | A kind of method of end-to-end QoS guarantee in SDN network | |
| Wang et al. | Implementation of multipath network virtualization with SDN and NFV | |
| CN114285758A (en) | Communication network optimization simulation system, method and device based on OPNET | |
| US20230198896A1 (en) | Method for communication in data center, apparatus, and system | |
| CN103067291A (en) | Method and device of up-down link correlation | |
| CN108400936A (en) | Information Network method for routing based on MPLS | |
| CN105472484A (en) | Wave channel balancing route wavelength allocation method of power backbone optical transport network | |
| Shen et al. | How powerful switches should be deployed: A precise estimation based on queuing theory | |
| CN103067294B (en) | Based on the method for the data flow equilibrium treatment of stream order-preserving in multi-next-hop forwarding router | |
| Fatmi et al. | Distributed multipath routing for data center networks based on stochastic traffic modeling | |
| Hertiana et al. | A joint approach to multipath routing and rate adaptation for congestion control in openflow software defined network | |
| Cheocherngngarn et al. | Depth-First Worst-Fit Search based multipath routing for data center networks | |
| Meng et al. | Efficient load balancing multipath algorithm for fiber-wireless network virtualization | |
| CN116647498A (en) | Dynamic selection method and system for multi-link network path | |
| Huang et al. | An enhanced scheduling framework for elephant flows in SDN-based data center networks | |
| CN111245724A (en) | SDN load balancing routing method based on virtual switch deployment | |
| Alanazi et al. | CAFT: Congestion-aware fault-tolerant load balancing for three-tier clos data centers | |
| Sharma et al. | An adaptive, fault tolerant, flow-level routing scheme for data center networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |