GB2181926A - Network data flow control - Google Patents
Network data flow control Download PDFInfo
- Publication number
- GB2181926A GB2181926A GB08622239A GB8622239A GB2181926A GB 2181926 A GB2181926 A GB 2181926A GB 08622239 A GB08622239 A GB 08622239A GB 8622239 A GB8622239 A GB 8622239A GB 2181926 A GB2181926 A GB 2181926A
- Authority
- GB
- United Kingdom
- Prior art keywords
- node
- speed
- data
- path
- buffer
- 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
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F5/00—Methods or arrangements for data conversion without changing the order or content of the data handled
- G06F5/06—Methods or arrangements for data conversion without changing the order or content of the data handled for changing the speed of data flow, i.e. speed regularising or timing, e.g. delay lines, FIFO buffers; over- or underrun control therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2205/00—Indexing scheme relating to group G06F5/00; Methods or arrangements for data conversion without changing the order or content of the data handled
- G06F2205/06—Indexing scheme relating to groups G06F5/06 - G06F5/16
- G06F2205/061—Adapt frequency, i.e. clock frequency at one side is adapted to clock frequency, or average clock frequency, at the other side; Not pulse stuffing only
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Communication Control (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
Data flow in a communications network is controlled by a downstream node specifying the maximum average speeds of a plurality of data streams sent from an upstream node on a plurality of data paths. As shown 2000 FIFO buffers 48 are dynamically allocated to the needs of the paths 40. The paths and local terminals 50 are connected to the node 42 via port protocol units 44,54 and input/output buffers 46,52. Paths are selected for speed change and upstream nodes informed by comparing previously specified speeds with the ideal speeds for the paths and taking account of buffer utilization and the times since speed increases and decreases were last considered. <IMAGE>
Description
SPECIFICATION
Network data flow control
This invention relates to controlling the flow of data among the nodes of a data communication network.
Two data terminals connected respectively to two different nodes of such a network communicate, in one typical scheme, via a socalled virtual circuit data path that passes through the two nodes and via a chain of intervening nodes and physical links that together complete the connection between the two data terminals. Each node or link may handle a large number of paths. The paths are multiplexed on the link. Each node has buffers that temporarily store inbound data of various paths until it can either be sent out along the outbound links of the respective paths or delivered to data terminals served by that node.
The average rate at which inbound data is received cannot exceed the rate of outbound data for very long without the buffers overflowing and data being lost.
One technique for controlling the flow of data in such a network is called an ARQ window. In such a scheme a node receiving data over a path. from an upstream node must periodically acknowledge receipt of blocks of data before the upstream node will send more. A maximum number of unacknowledged frames (the ARQ window) can be on the path at one time. The receiving node can thus regulate the quantity of data received over a given path by acknowledging or not acknowiedging incoming frames.
A general feature of the invention is controlling the flow of data in a communications network of the kind in which a plurality of data streams are sent respectively via a plurality of data paths from an upstream node to a downstream node, by having the downstream node specify the maximum average speeds of data streams on the respective paths. As a result each node can directly dictate the speeds of data streams on inbound data paths, reducing loss of data.
Preferred embodiments of the invention include the following features. The downstream node has buffer means for temporarily storing the data streams, the current utilization of the buffer means is measured, and the speeds are specified based on the utilization. A determination is made of the time when speed changes should be considered, and data paths whose speeds are to be changed are identified, thus enabling the necessary changes in the aggregate inbound data rate to be made on the most appropriate paths. The decision of when to consider speed changes itself depends on buffer utilization and on how long ago speed changes were considered; thus changes are considered only as frequently as is appropriate in view of how heavily the buffer is being used.The buffer means includes a plurality of buffers and the buffer utilization is measured as a percentage of the buffers that are being used, a simple and highly relevant criterion. Increases and decreases in speed are considered separately based on buffer utilization and on how long ago speed increases or decreases respectively were considered, thus enabling some path speeds to be increased and others to be decreased at the same time to efficiently manage the buffer utilization.Speed increases or decreases are considered when the previous consideration of speed increases or decreases respectively occurred longer ago than a predetermined amount of time (whose magnitude varies directly with utilization of the buffer, in the case of the decision to consider speed increases, and inversely with the buffer utilization for speed decreases); thus speed increases are considered infrequently when the buffer utilization is high and speed decreases are considered infrequently when the buffer utilization is low. A data path is picked for a speed increase or decrease based on whether the ratio of that path's previously specified speed to a so-called ideal speed exceeds respectively a high or low threshold that varies with buffer utilization.Each threshold increases with higher buffer utilization, so that the higher the buffer utilization the smaller the previously specified speed must be relative to the ideal speed in order to trigger a speed change. The threshold for speed decreases never exceeds unity; the threshold for speed increases never falls below unity; thus speed decreases will never be triggered for a path unless its most recently specified speed is above the ideal speed. The new speed for a path is equal to the ideal speed times a factor no smaller than unity, the factor being a value that varies inversely with buffer utilization; as a result, for very low utilization, the new speed will be higher than the ideal speed.The ideal speed is based on the previous speed set for the path by a node further downstream, on the maximum bandwidth available for the path, and on the utilization of the buffers allocated to a given path. If that utilization exceeds two minutes worth of data bytes at the unadjusted ideal speed, the ideal speed is reduced. The downstream node communicates the speeds to the upstream node, thus enabling the downstream node to effectively control its own buffer utilization.
Other advantages and features will become apparent from the following description of the preferred embodiment, and from the claims.
We first briefly describe the drawings, in which:
Fig. 1 is a block diagram of a data communication network;
Fig. 2 is a format diagram of a scan;
Fig. 3 is a block diagram of a node;
Fig. 4 is a block diagram of an input/output processor of a node;
Fig. 5 is a table of maximum timer intervals for various buffer utilizations;
Fig. 6 is a flow control table; and
Fig. 7 is a format diagram of a set speed message.
Referring to Fig. 1, data communication network 10 includes nodes 12, 14, 16, 18, 20 interconnected by data links 22, 24, 26, 28, 30. Each node serves one or more data terminals 32. Data flows from a first terminal to a second terminal over a preestablished virtual circuit data path that includes the nodes which serve the two terminals and some combination of intervening links and nodes necessary to complete a physical connection between the two terminals. For example, a path may be established from terminal 33 to terminal 35 via node 18, link 26, node 16, link 28, and node 20. Data paths are created or terminated from time to time according to need. Multiple data paths are multiplexed onto each physical data link.
Referring to Fig. 2, to accomplish the multiplexing each link carries a succession of scans. Each scan is a series of 8-bit bytes arranged in fields including a scan initiator field 34 which uniquely marks the beginning of the scan. The remaining fields of the scan alternate between two types called deltas 36 and slots 38. Each scan potentially includes a slot 38 for each path currently established on that link. For example, slot 1 in successive scans together make up a particular path on the link and can carry a stream of data bytes.
Each slot has a variable number of data bytes, so that the relative maxiumum average rates at which data bytes of different paths can be carried on the link is determined by the relative maximum numbers of data bytes (called slot weights) in the respective slots.
The absolute rates at which data streams can be carried on a path also depend on the total length of each scan and the bit rate of the link. The speeds at which various paths can carry data can be changed by the upstream node reconfiguring the relative slot weights.
The scan rate is no higher than a predetermined maximum rate and the scan rate changes as the slot weights change. Each slot has a maximum slot weight.
Although a given path has a certain speed at a particular time, the upstream terminal for the path may not be using the path at that time. If, when a scan is being assembled for transmission, an established path is not in use the slot corresponding to that path is omitted from the scan to save bandwidth. Each delta indicates how many, if any, slots have been omitted between two consecutive slots that are in the scan.
Referring to Fig. 3, a link 40 is connected to a node 42 via a link port 44 and an I/O buffer 46. The scans are carried on link 40 using an HDLC link layer protocol that is handled by link port 44 so that the HDLC protocol is transparent to buffer 46, which deals only with the bytes of the scan. Buffer 46 serves as a temporary buffer for incoming and outgoing data bytes. Node 42 includes an input/output processor (IOP) 46 that gets each incoming byte from buffer 46 and, in the case of a data byte of a slot, puts the data byte into a 32-byte long FIFO data buffer 48 that has been assigned to the path corresponding to that slot. A pool of 2000 FIFO data buffers 48 together make up a node buffer pool 49, which has a finite capacity. Data buffers 48 are allocated dynamically to the needs of the respective paths.If ail of the buffers allocated to a given path are full when another data byte arrives on that path, another buffer, if available, is allocated to that path. The addresses of the buffers presently allocated to a path are kept as a linked list.
Periodically, the IOP removes data bytes waiting in the respective data buffers. In the case of data bytes intended for delivery to another node, the IOP inserts the data bytes into the appropriate slots of scans and delivers the scans to the appropriate link port (only one of several of which are shown in Fig. 3).
Data bytes intended for a terminal 50 served by the node are passed via a buffer 52 and a terminal port 54 which handles the protocol used between the node and the terminal. If removing a data byte from a given buffer 48 leaves that buffer empty, it becomes freed up for reallocation to another path.
The flow of data bytes into node 42 via links from upstream nodes is regulated in the following way in order to effectively use the available capacity of buffers 48 without unnecessarily discarding data.
Referring to Fig. 4, IOP 46 contains a data buffer utilization monitor 58 which every two seconds looks at node buffer 49 and determines buffer utilization as the percentage of buffers 48 that are currently not empty. Based in part on this determination, a speed up/slow down evaluator 60 decides whether the individual paths should be examined and a number of them selected for which the incoming data rate from the upstream node should be slowed, or accelerated, or no action should be taken. A speed up timer 62 and a slow down timer 64 respectively keep track of the amounts of time which have elapsed since the last time consideration was given respectively to speeding up or slowing down the inbound data speed on any path. A maximum interval table 66 in evaluator 60 relates the percentage of buffers 48 currently being used (i.e., not empty) to the maximum elapsed times on timers 62, 64 before speeding up or slowing down should be considered.
Referring to Fig. 5, maximum interval table 66 contains, for each successive range of 10 percent in the buffer used percentage, the maximum intervals which may elapse on each of timers 62, 64 without analyzing paths to determine whether they should be sped up or slowed down, respectively. For example, given a buffer filled percentage of 35 percent, if the value of the speed up timer exceeds three seconds an analysis will be made of each path as to whether its inbound data rate, expressed in bytes per second, should be increased. If the values of both timers exceed the intervals in their respective columns in table 66, both analyses are made for each path.
The maximum interval for the speed up timer increases with higher buffer used percentages, and conversely for the slow down timer.
Therefore, when the buffer is almost full, speeding up is considered relatively infrequently while slowing down is considered relatively frequently.
Referring again to Fig. 4, once a decision to evaluate paths has been made, evaluator 60 triggers a path analyzer 65 which considers individual paths for speeding up or slowing down as the case may be, based on the path's ideal speed and a flow control table 68.
Path analyzer 65 includes an ideal speed calculator 70 that calculates an ideal speed 67 for each path based on the most recently specified (current) speed 72 specified for data leaving the node on an outbound leg of the path as found in current speed buffers 71, and the maximum outbound bandwidth 74 available to the path, derived from maximum bandwith buffers 75.
One of two formulas for the ideal speed is applied by calculator 70; the determination of which formula will be used is based on whether or not more than two seconds worth of outgoing data for that path is currently buffered. Ideal speed calculator 70 watches the utilization of each path's buffers 48. While the path's buffers hold less than two seconds worth of outgoing data at the unadjusted ideal speed, the ideal speed is updated to the highest speed at which data can be transmitted to the down stream node, i.e., ideal speed equals the minimum of the previously set speed or the maximum bandwidth.
When more than two seconds worth of outgoing data for a path (at the unadjusted ideal speed) has accumulated, data is being received faster than node 42 can retransmit it to a downstream node or deliver it to a terminal 50. It is therefore desirable to reduce the rate at which data arrives for that path at node 42 by 50 percent. In that case ideal speed 67 is updated to be equal to one-half of the minimum of the current rate or the maximum banwidth.
Once the ideal speed 67 has been updated, calculator 70 calculates a speed ratio 76 which is the ratio of the updated ideal speed to the most recent inbound speed 72. If this ratio is less than 1.0, it indicates that data can be accepted from the upstream node at a higher rate; if the ratio is greater than 1.0, it indicates that data is arriving at too great a rate.
Referring to Fig. 6, flow control table 68 is a static table which, based on the speed ratio 76 and the buffers used percentage 77, specifies whether the path associated with a given speed ratio should be selected for speeding up or slowing down and, if it is selected, provides a change factor 80 by which the ideal speed for the given path is multiplied to arrive at the new incoming speed for that path. Two columns 100, 101 contain threshold ratio values for use respectively in making speed up or slow down decisions. The low threshold values 100 are used when selecting paths to slow down. If for a given buffers used percentage, the speed ratio 76 is less than the corresponding low threshold value 100, the path will be selected, otherwise it will not. As the buffer filled percentage decreases, the speed ratio must be increasingly smaller to trigger an incoming data flow reduction.For example, a speed ratio of .9 will cause a path to be selected for slowing down if the buffers are 80% used. If, however, the buffers are only 8% used, the speed ratio must be less than .05 to trigger slowing down.
The high threshold values 101 are applied when paths to be sped up are being selected.
If a given speed ratio is higher than the high threshold level indicated for the current buffers used percentage the path will be selected. As the buffers used percentage increases, the speed ratio must be increasingly greater than 1.0 for the path to be selected.
Referring again to Fig. 4, to obtain the proper change factor, calculator 70 applies the speed ratio 76 to storage 68 which then passes to a new speed calculator 110 either the indicated change factor or a null value. A change factor indicates that the path currently under evaluation has been selected for a speed change. A null value indicates that the path has not been selected. If a change factor is received, it is multiplied by ideal speed 67 to obtain a new speed 102.
A final check is done on each path before it is selected to be certain that it is active. Data buffer utilization monitor 58 provides another value, the thirty second average 104, expressed in bytes per second, transmitted over the path. A non-zero average 104 indicates that the path is active. New speed calculator 110 examines average 104 and, if it is nonzero, selects that path by delivering a path identifier and a new speed value to a set speed assembler 120.
Each path is evaluated in the above described manner and, for those selected, the new speed 102 is calculated. These speeds 102, which are incoming data rates for each path, are communicated to the upstream node by set speed assembler 120. The upstream node effectuates the new speeds by adjustments to the slot weights, i.e. the maximum number of data bytes included in each path's slot.
To communicate the desired rates to the upstream node, a set speed assembler 120 assembles the desired rates into a special message called a set speed 122.
Referring to Fig. 7, a set speed 122 contains for each selected path a path identifier (path id) 124 and the new speed 102 for that path. Each set speed is broken into fields which alternate between two types; the path ids 124 occupy the first type of field and the new speeds 102 occupy the second.
Set speeds are sent to the upstream node over a control path that is carried on the link.
Each upstream node disassembles each received set speed 122, stores the new speeds 102 in its current speed buffers 71, and reconfigures its outbound scans accordingly.
Claims (23)
1. A node adapted for incorporation in a communications network of the kind in which a plurality of data streams are sent respectively via a plurality of data paths from an upstream node to a downstream node as a downstream node, said node being provided with means adapted to control the flow of data by specifying the maximum average speeds of said data streams on the respective paths.
2. A node according to Claim 1, wherein said data flow control means comprise: means for determining when changes in said speeds will be considered, and means for identifying data paths whose speeds are to be changed.
3. A node according to Claims 1 or 2, further comprising buffer means for temporarily storing said data streams, said speeds being specified based on the utilization of said buffer means.
4. A node according to Claim 3, wherein said buffer means comprises a plurality of buffers and said data flow control means includes means for measuring said buffer utilization as a percentage of said buffers which are being used.
5. A node according to both Claim 2 and any of Claims 3 or 4, wherein said determining means bases said determination on said buffer utilization and on how long ago speed changes were previously considered.
6. A node according to Claim 2 or any claim appendent thereto wherein said determining means comprises: means for deciding when increases in said speeds should be considered, said decision being based on said buffer utilization and on how long ago speed increases were previously considered; and means for deciding when decreases in said speeds should be considered, said decision being based on said buffer utilization and on how long ago speed decreases were previously considered.
7. A node according to both Claim 5 and
Claim 6, wherein increases or decreases in said speed are considered when the previously considered speed increases or decreases respectively occurred longer ago than a predetermined amount of time whose magnitude depends on the utilization of said buffer.
8. A node according to Claim 7, wherein said predetermined amount of time is larger for higher buffer utilization in the case of the decision to consider speed increases, and is lower for higher buffer utilization in the case of the decision to consider speed decreases.
9. A node according to both Claim 2 and any one of Claims 3 or 4, wherein said path identifying means comprise: means for comparing the previously specified speed of each said path with an ideal speed for said path, and means for selecting a data path for a speed change based on said comparison and on said buffer utilization.
10. A node according to Claim 9, wherein said comparison is a ratio and said path is selected for a speed change if said ratio exceeds a threshold that varies with the utilization of said buffer.
11. A node according to Claim 10, wherein said threshold increases with higher utilization.
12. A node according to Claims 10 or 11, wherein there is one said threshold that never exceeds unity for speed decreases and a second said threshold that is never less than unity for speed increases.
13. A node according to any of Claims 9 to 12, wherein said data flow control means further comprises means for establishing a new speed equal to said ideal speed times a factor no smaller than unity.
14. A node according to Claim 13, wherein said factor depends on the buffer utilization and higher said factors are associated with lower said utilizations.
15. A node according to Claim 9, wherein said means for comparing bases said ideal speed on the previous speed set for said path by a node further downstream from from said downstream node and on the maximum bandwidth available for said path.
16. A node according to Claim 9, wherein said buffer means comprises a plurality of buffers allocated to respective said paths and said means for comparing further bases said ideal speed on the utilization of said buffers allocated to a given said path.
17. A node according to Claim 16, wherein said ideal speed is reduced when the utilization of said allocated buffers exceeds a maximum.
18. A node according to Claim 17, wherein said maximum comprises said allocated buffers containing two seconds worth of data bytes at said ideal speed prior to said reduction.
19. A node according to any preceding claim, further provided with means for communicating said speeds from said node to said upstream node.
20. A node substantially as hereinbefore described with reference to and as shown in the accompanying drawings.
21. A method for controlling the flow of data in a communications network of the kind in which a plurality of data streams are sent respectively via a plurality of data paths from an upstream node to a downstream node by causing the downstream node to specify the maximum average speeds of said data streams on the respective paths.
22. A communications network of the kind in which a plurality of data streams are sent respectively via a plurality of data paths from an upstream node to a downstream node, wherein each node in said network adapted to serve as a downstream node comprises a node according to any of Claims 1 to 20.
23. A method for controlling the flow of data in a communications network substantially as hereinbefore described with reference to the accompanying drawings.
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US77707885A | 1985-09-17 | 1985-09-17 |
Publications (3)
Publication Number | Publication Date |
---|---|
GB8622239D0 GB8622239D0 (en) | 1986-10-22 |
GB2181926A true GB2181926A (en) | 1987-04-29 |
GB2181926B GB2181926B (en) | 1990-02-14 |
Family
ID=25109233
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
GB8622239A Expired - Lifetime GB2181926B (en) | 1985-09-17 | 1986-09-17 | Network data flow control |
Country Status (3)
Country | Link |
---|---|
JP (1) | JPH0771124B2 (en) |
AU (1) | AU591702B2 (en) |
GB (1) | GB2181926B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2242097A (en) * | 1990-01-19 | 1991-09-18 | British Broadcasting Corp | Buffer synchronisation in a variable transmission rate codec |
GB2274041A (en) * | 1990-01-19 | 1994-07-06 | British Broadcasting Corp | Buffer Resynchronisation in a variable transmission rate coder |
WO1995011557A1 (en) * | 1993-10-23 | 1995-04-27 | International Business Machines Corporation | Selective congestion control mechanism for information networks |
GB2337403A (en) * | 1998-05-11 | 1999-11-17 | Gen Datacomm Adv Res | Data communication |
GB2389022A (en) * | 2002-05-22 | 2003-11-26 | 3Com Corp | Network device with selectable data rate based on traffic volume |
WO2005124556A2 (en) * | 2004-06-14 | 2005-12-29 | Koninklijke Philips Electronics N.V. | Interface device for debugging and/or tracing a computer system comprising one or multiple masters and one or multiple slaves working together. |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH02228641A (en) * | 1989-02-28 | 1990-09-11 | Kokichi Tateno | Device for forming format document |
US5029164A (en) * | 1990-04-13 | 1991-07-02 | Digital Equipment Corporation | Congestion avoidance in high-speed network carrying bursty traffic |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2022373B (en) * | 1978-04-28 | 1983-02-02 | Ericsson L M Pty Ltd | Interconntecting terminal device in a digital data network |
DE3470266D1 (en) * | 1983-12-27 | 1988-05-05 | American Telephone & Telegraph | Communication system having reconfigurable data terminals |
JPS60171849A (en) * | 1984-02-17 | 1985-09-05 | Hitachi Ltd | Data low control system |
-
1986
- 1986-09-04 AU AU62448/86A patent/AU591702B2/en not_active Expired
- 1986-09-17 GB GB8622239A patent/GB2181926B/en not_active Expired - Lifetime
- 1986-09-17 JP JP61219109A patent/JPH0771124B2/en not_active Expired - Lifetime
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2242097A (en) * | 1990-01-19 | 1991-09-18 | British Broadcasting Corp | Buffer synchronisation in a variable transmission rate codec |
GB2274041A (en) * | 1990-01-19 | 1994-07-06 | British Broadcasting Corp | Buffer Resynchronisation in a variable transmission rate coder |
GB2274041B (en) * | 1990-01-19 | 1994-09-14 | British Broadcasting Corp | Buffer regulation |
GB2242097B (en) * | 1990-01-19 | 1994-09-14 | British Broadcasting Corp | Buffer resynchronisation |
WO1995011557A1 (en) * | 1993-10-23 | 1995-04-27 | International Business Machines Corporation | Selective congestion control mechanism for information networks |
US5768258A (en) * | 1993-10-23 | 1998-06-16 | International Business Machines Corporation | Selective congestion control mechanism for information networks |
GB2337403A (en) * | 1998-05-11 | 1999-11-17 | Gen Datacomm Adv Res | Data communication |
GB2389022A (en) * | 2002-05-22 | 2003-11-26 | 3Com Corp | Network device with selectable data rate based on traffic volume |
WO2005124556A2 (en) * | 2004-06-14 | 2005-12-29 | Koninklijke Philips Electronics N.V. | Interface device for debugging and/or tracing a computer system comprising one or multiple masters and one or multiple slaves working together. |
WO2005124556A3 (en) * | 2004-06-14 | 2006-11-30 | Koninkl Philips Electronics Nv | Interface device for debugging and/or tracing a computer system comprising one or multiple masters and one or multiple slaves working together. |
US7660963B2 (en) | 2004-06-14 | 2010-02-09 | Nxp B.V. | Interface device for debugging and/or tracing a computer system comprising one or multiple masters and one or multiple slaves working together |
Also Published As
Publication number | Publication date |
---|---|
GB2181926B (en) | 1990-02-14 |
AU6244886A (en) | 1987-03-19 |
AU591702B2 (en) | 1989-12-14 |
JPH0771124B2 (en) | 1995-07-31 |
JPS62104245A (en) | 1987-05-14 |
GB8622239D0 (en) | 1986-10-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US4901277A (en) | Network data flow control technique | |
EP0577359B1 (en) | Congestion control in high speed networks | |
EP0732019B1 (en) | Control of overload situations in frame relay network | |
EP0464024B1 (en) | Congestion free packet network | |
CA2130403C (en) | Atm switching system and cell control method | |
EP0587522B1 (en) | Rate-based congestion control in packet communications networks | |
KR100234212B1 (en) | The congestion control based on adaptive speed control in packet network | |
EP0591150B1 (en) | Congestion management based on multiple framing strategy | |
EP0736238B1 (en) | Data link interface for packet-switched network | |
EP1508227B1 (en) | Buffer memory reservation | |
US5970048A (en) | Control of overload situations in frame relay network | |
US5822540A (en) | Method and apparatus for discarding frames in a communications device | |
US6963578B2 (en) | Router | |
US6539026B1 (en) | Apparatus and method for delay management in a data communications network | |
US6452901B1 (en) | Cell loss balance system and method for digital network | |
EP0920235A2 (en) | Congestion management in a multi-port shared memory switch | |
JPH0657016B2 (en) | Congestion control type packet switching method and apparatus thereof | |
US7839787B2 (en) | Delay-based overflow routing in communication systems | |
GB2181926A (en) | Network data flow control | |
US6912224B1 (en) | Adaptive playout buffer and method for improved data communication | |
KR100745679B1 (en) | Method and apparatus for packet scheduling using adaptation round robin | |
US6785235B1 (en) | Priority control of queued data frames in frame delay multiplexing | |
JP4104756B2 (en) | Method and system for scheduling data packets in a telecommunications network | |
EP0872086A1 (en) | Method and apparatus for discarding frames in a communications device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PE20 | Patent expired after termination of 20 years |
Effective date: 20060916 |