CN111683017A - Multi-level congestion control method, device, system and medium in high-speed interconnection network - Google Patents
Multi-level congestion control method, device, system and medium in high-speed interconnection network Download PDFInfo
- Publication number
- CN111683017A CN111683017A CN202010456010.1A CN202010456010A CN111683017A CN 111683017 A CN111683017 A CN 111683017A CN 202010456010 A CN202010456010 A CN 202010456010A CN 111683017 A CN111683017 A CN 111683017A
- Authority
- CN
- China
- Prior art keywords
- congestion
- message
- value
- mark
- cnp
- 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
- H04L47/12—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
- H04L47/11—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
- H04L47/25—Flow control; Congestion control with rate being modified by the source upon detecting a change of network conditions
-
- 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 discloses a method, a device, a system and a medium for controlling multi-level congestion in a high-speed interconnection network. The invention reads the value of the congestion mark carried in the CNP message through the source node, adjusts the rate of sending the message according to the value of the congestion mark to reduce the message congestion, and the value of the congestion mark is a numerical value and is obtained by accumulating each exchange node when the message is transmitted to be congested, thereby being capable of sensing the congestion propagation condition, congestion level and the like of different paths, dynamically controlling the message sending rate or adjusting the route and the like by utilizing the information, and further realizing more accurate congestion control.
Description
Technical Field
The invention relates to a high-speed interconnection network technology for a high-performance computer, in particular to a method, a device, a system and a medium for controlling multi-level congestion in a high-speed interconnection network.
Background
High-speed interconnect networks usually employ credit-based flow control mechanisms (such as InfiniBand), which have the advantages of high transmission rate, low transmission delay, no lost packets, and the like, and are therefore widely used in high-performance computing systems. However, when network congestion occurs when the load of the switching or destination nodes exceeds the maximum processing capacity, the credit control mechanism is more likely to cause congestion propagation, thereby reducing the overall performance of the network. In large-scale high-speed network systems, congestion becomes an important factor affecting network performance.
To solve the congestion problem, a congestion control mechanism based on Explicit Congestion Notification (ECN) is generally adopted in the high-speed interconnection network, and the generation of a congestion tree is prevented by detecting congestion in the network and explicitly notifying the source node, so as to reduce the rate of data injection into the network. The congestion control mechanism comprises the following steps:
1. when a message sent by a source node passes through a switching node, the switching node judges the occupation of a buffer area, the credit of the next level and the like to detect Congestion, and if the Congestion is detected, the corresponding domain in a message header is marked with Congestion, so that the message carries 1-bit Forward Explicit Congestion notification FECN (Forward Explicit Congestion notification fi, FECN mark for short).
2. When a destination node receives a Congestion marked message (carrying a forward Explicit Congestion notification FECN), a Congestion notification message (Congestion notification fi Packet, abbreviated as a CNP message) is immediately generated and sent to a source node, which is called a Backward Explicit Congestion Notification (BECN);
3. after receiving the CNP message, the source node adjusts the congestion Control Table index CCTI (congestion Control Table index) of the corresponding port or virtual channel vc (virtual channel) according to a certain rule, so as to adjust the CCTI and Control the rate of sending the message, thereby achieving the purpose of relieving the network congestion.
The congestion control mechanism adjusts the message injection rate in a mode of matching the switching node, the destination node and the source node, can effectively relieve the problem of network performance reduction caused by congestion, can better utilize the network bandwidth, improves the fairness and reduces the delay of the Victim flow. However, the congestion control mechanism simply performs congestion marking on all the packets passing through the congestion path, and all the packets passing through the congestion path use the same marking mode, and does not consider the propagation situation of congestion in the switching network, and cannot accurately distinguish the severity of congestion propagation on different paths, so that different strategies cannot be adopted for various types of flows when the sending end performs packet sending rate control, and congestion control may be inaccurate.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: the invention provides a multi-level congestion control method, a device, a system and a medium in a high-speed interconnection network, aiming at the defects that the congestion propagation condition in the congestion control mechanism of the existing high-speed interconnection network is simpler, and the congestion propagation condition in the network cannot be marked and distinguished, so that different rate control strategies cannot be adopted for various flows, and the congestion control is inaccurate.
In order to solve the technical problems, the invention adopts the technical scheme that:
a multi-level congestion control method in a high-speed interconnection network, a source node in the high-speed interconnection network processes a CNP message, and the steps comprise:
s1) receiving the CNP message returned by the destination node;
s2) reading the value of the congestion mark carried in the CNP message, wherein the value of the congestion mark is a numerical value and is obtained by accumulating when each exchange node transmits the message and congestion occurs;
s3) adjusting the message sending rate according to the value of the congestion mark to reduce the message congestion.
Optionally, the detailed step of step S3) includes:
s3.1) acquiring the corresponding CCTI stepping length according to the value of the congestion mark;
s3.2) updating the CCTI value according to the CCTI stepping length;
s3.3) controlling the message sending rate according to the CCTI value to reduce message congestion.
Optionally, step S1) is preceded by a step of sending a CNP message by a destination node in the high-speed interconnection network:
t1) reading the value of the congestion mark in the message header slice of the received message, and skipping to execute the next step when the value of the congestion mark in the message header slice of the message is larger than a preset initial value;
t2) generates a CNP message, and adds the value of the congestion flag in the header slice of the received message to the CNP message, and transmits the value of the congestion flag in the header slice of the received message back to the source node through the CNP message.
Optionally, the step T2) of attaching the value of the congestion flag in the header slice of the received packet to the CNP packet specifically means that the value of the congestion flag in the header slice of the received packet is attached to a multi-bit field specified in the header slice of the CNP packet.
Optionally, the step T1) of forwarding the packet by a switching node in the high-speed interconnection network further includes:
E1) receiving a message from the previous hop;
E2) judging whether the congestion condition occurs at present, and if the congestion condition does not occur, skipping to execute the step E3); otherwise, accumulating the congestion mark value in the message header slice of the message;
E3) and outputting the message to the next hop.
Optionally, the detailed step of accumulating the values of the congestion flags in the header slice of the packet in step E2) includes: reading the value of the congestion mark in the message header slice of the message, judging whether the value of the congestion mark is equal to the maximum value which can be stored by the congestion mark, if the value of the congestion mark is equal to the maximum value which can be stored by the congestion mark, keeping the value of the congestion mark in the message header slice of the message unchanged, and skipping to execute the step E3); otherwise, adding 1 to the value of the congestion mark on the basis of the original value, updating the congestion mark in the message header slice of the message, and skipping to execute the step E3).
In addition, the present invention also provides a multi-level congestion control device in a high-speed interconnection network, comprising:
a CNP message receiving program unit, configured to receive a CNP message returned by a destination node;
a CNP message analysis program unit, configured to read a value of a congestion flag carried in a CNP message, where the value of the congestion flag is a numerical value and is obtained by accumulating congestion occurring in a transmission message through each switching node;
and the message rate control program unit is used for adjusting the rate of sending the message according to the value of the congestion mark so as to reduce the message congestion.
Furthermore, the present invention also provides a multi-level congestion control device in a high-speed interconnection network, the multi-level congestion control device being programmed or configured to execute the steps of the multi-level congestion control method in the high-speed interconnection network, or a computer program or a firmware program being programmed or configured to execute the multi-level congestion control method in the high-speed interconnection network being stored in a memory of the multi-level congestion control device.
In addition, the invention also provides a high-speed interconnection network system, which comprises a source node, a destination node and a switching node, wherein the source node passes through at least one switching node and one destination node, the source node is programmed or configured to execute the steps of the multi-level congestion control method in the high-speed interconnection network, or the destination node is programmed or configured to execute the steps of the multi-level congestion control method in the high-speed interconnection network, or the switching node is programmed or configured to execute the steps of the multi-level congestion control method in the high-speed interconnection network.
Furthermore, the present invention also provides a computer-readable storage medium having stored therein a computer program or firmware program programmed or configured to execute the method of multi-level congestion control in the high speed interconnect network.
Compared with the prior art, the invention has the following advantages:
1. the invention reads the value of the congestion mark carried in the CNP message through the source node, adjusts the rate of sending the message according to the value of the congestion mark to reduce the message congestion, and the value of the congestion mark is a numerical value and is obtained by accumulating each exchange node when the message is transmitted to be congested, thereby being capable of sensing the congestion propagation condition, congestion level and the like of different paths, dynamically controlling the message sending rate or adjusting the route and the like by utilizing the information, and further realizing more accurate congestion control.
2. The method only needs to transmit the value of the congestion mark in the transmission process, and the resource consumption is low.
3. The Congestion control mechanism based on the existing Explicit Congestion Notification (ECN) only needs to expand the message and the CNP message to carry the value of the Congestion mark and change the related control method, only needs to slightly change the existing ECN-based Congestion control mechanism, does not change other software and hardware of the existing network switching chip, and does not affect the performance, the implementation method and the implementation cost.
Drawings
Fig. 1 is a schematic flow chart of a source node processing a CNP packet in the method according to the embodiment of the present invention.
Fig. 2 is a schematic diagram of a coordination principle of different nodes in the embodiment of the present invention.
Fig. 3 is a detailed flowchart of different nodes in the embodiment of the present invention.
Detailed Description
As shown in fig. 1, fig. 2 and fig. 3, in the method for controlling multi-level congestion in a high-speed interconnection network according to this embodiment, the step of processing a CNP packet by a source node in the high-speed interconnection network includes:
s1) receiving the CNP message returned by the destination node;
s2) reading the value of the congestion mark carried in the CNP message, wherein the value of the congestion mark is a numerical value and is obtained by accumulating when each exchange node transmits the message and congestion occurs;
s3) adjusting the message sending rate according to the value of the congestion mark to reduce the message congestion.
Because the value of the congestion mark is a numerical value and is obtained by accumulating when each switching node transmits the message and congestion occurs, when the message passes through a plurality of congested switching nodes, the congestion mark is increased until the mark is maximum when each message passes through one congested node, so that the congestion mark represents the congestion propagation condition on the path through which the message passes, and the congestion propagation is longer when the value is larger. On the basis, the source node can effectively reduce the message congestion by adjusting the message sending rate according to the value of the congestion mark.
As an optional manner, the step S2) or S3) further includes a step of generating a management packet from the value of the congestion flag carried in the CNP packet, sending the management packet to the management node, and reporting the severely congested path information for the management server to perform system-level load analysis, routing control, network optimization, and the like.
As shown in fig. 2 and 3, the detailed step of step S3) in the present embodiment includes:
s3.1) acquiring the corresponding CCTI stepping length according to the value of the congestion mark; the size of the value of the congestion mark and the mapping mode between the CCTI step lengths are preset, and different processing modes such as a mapping table, a function, a Hash mapping and the like can be adopted according to needs as long as the mapping between the size of the value of the congestion mark and the CCTI step lengths can be realized;
s3.2) updating the CCTI value according to the CCTI stepping length;
s3.3) controlling the message sending rate according to the CCTI value to reduce message congestion.
It should be noted that, the conventional usage method for updating the CCTI value according to the CCTI step length and controlling the rate of sending the packet according to the CCTI value to reduce the packet congestion to CCTI is not described herein again.
As shown in fig. 1, fig. 2, and fig. 3, before step S1), this embodiment further includes a step of sending a CNP packet by a destination node in the high-speed interconnection network:
t1) reading the value of the congestion mark in the message header slice of the received message, and skipping to execute the next step when the value of the congestion mark in the message header slice of the message is larger than a preset initial value;
t2) generates a CNP message, and adds the value of the congestion flag in the header slice of the received message to the CNP message, and transmits the value of the congestion flag in the header slice of the received message back to the source node through the CNP message.
In this embodiment, the step T2) of attaching the value of the congestion flag in the header slice of the received packet to the CNP packet specifically means that the value of the congestion flag in the header slice of the received packet is attached to a multi-bit field specified in the header slice of the CNP packet. In this embodiment, adding the value of the congestion flag in the header slice of the received packet to the CNP packet specifically means that the BECN domain of the extended CNP packet is multi-bit (in this embodiment, the extended domain is named as a multi-level BECN flag domain, and the stored multi-level BECN flag is the congestion flag in the foregoing, as shown in fig. 3), and adding the value of the congestion flag to the extended BECN domain, which causes little change to the existing CNP packet, thereby enabling the existing CNP packet to have good compatibility with the existing CNP packet.
As shown in fig. 1, fig. 2, and fig. 3, before the step T1), the step of forwarding the packet by the switching node in the high-speed interconnection network further includes:
E1) receiving a message from the previous hop;
E2) judging whether the congestion condition occurs at present, and if the congestion condition does not occur, skipping to execute the step E3); otherwise, accumulating the congestion mark value in the message header slice of the message;
E3) and outputting the message to the next hop.
In this embodiment, the detailed step of accumulating the congestion flag value in the header slice of the packet in step E2) includes: reading the value of the congestion mark in the message header slice of the message, judging whether the value of the congestion mark is equal to the maximum value which can be stored by the congestion mark, if the value of the congestion mark is equal to the maximum value which can be stored by the congestion mark, keeping the value of the congestion mark in the message header slice of the message unchanged, and skipping to execute the step E3); otherwise, adding 1 to the value of the congestion mark on the basis of the original value, updating the congestion mark in the message header slice of the message, and skipping to execute the step E3). In this embodiment, the congestion flag is obtained by expanding the FECN bit in the existing ECN-based congestion control mechanism to multiple bits (in this embodiment, the expanded domain is named as a multi-level FECN flag domain, and the stored multi-level FECN flag is the congestion flag in the foregoing, as shown in fig. 3), so that the congestion propagation condition on the path through which the message passes can be recorded by the congestion flag obtained after expansion, and the information can be used for more accurate message sending rate control; in addition, the congestion flag may also be reported to a network system management node through a switching node, or a destination node, or a source node, for system-level load analysis, routing control, network optimization, and the like.
As an optional implementation manner, in this embodiment, when the source node sends a packet, a value of a congestion flag carried in the packet is initially 0, and in addition, other initialization values may also be adopted as needed.
In summary, the multi-level congestion control method in the high-speed interconnect network of this embodiment only needs to make minor changes (mainly including an FECN domain for extending packets, a BECN domain for extending CNP packets, a modified packet sending rate control logic, or a management packet generation logic) on the existing congestion control mechanism based on the ECN, and does not change other software and hardware of the existing network switch chip, and does not affect the performance, the implementation method, and the implementation cost. The multi-level congestion control method in the high-speed interconnection network of the embodiment enables the multi-level congestion control method to carry more original network information by a method of expanding congestion marks, so that the congestion can be relieved by accurately distinguishing and controlling congestion propagation conditions on different paths in the network.
In addition, the present embodiment further provides a multi-level congestion control device in a high-speed interconnection network, including:
a CNP message receiving program unit, configured to receive a CNP message returned by a destination node;
a CNP message analysis program unit, configured to read a value of a congestion flag carried in a CNP message, where the value of the congestion flag is a numerical value and is obtained by accumulating congestion occurring in a transmission message through each switching node;
and the message rate control program unit is used for adjusting the rate of sending the message according to the value of the congestion mark so as to reduce the message congestion.
In addition, the present embodiment also provides a multi-level congestion control device in a high-speed interconnection network, where the multi-level congestion control device is programmed or configured to execute the steps of the multi-level congestion control method in the high-speed interconnection network, or a computer program or a firmware program programmed or configured to execute the multi-level congestion control method in the high-speed interconnection network is stored in a memory of the multi-level congestion control device.
In addition, the present embodiment also provides a high-speed interconnection network system, which includes a source node, a destination node and a switch node, where the source node passes through at least one switch node and one destination node, and the source node is programmed or configured to execute the steps of the aforementioned multi-level congestion control method in the high-speed interconnection network, or the destination node is programmed or configured to execute the steps of the aforementioned multi-level congestion control method in the high-speed interconnection network, or the switch node is programmed or configured to execute the steps of the aforementioned multi-level congestion control method in the high-speed interconnection network.
Furthermore, the present embodiment also provides a computer-readable storage medium in which a computer program or a firmware program programmed or configured to execute the foregoing method for controlling congestion in multiple levels in a high-speed interconnection network is stored.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein. The present application is directed to methods, apparatus (systems), and computer program products according to embodiments of the application wherein instructions, which execute via a flowchart and/or a processor of the computer program product, create means for implementing functions specified in the flowchart and/or block diagram block or blocks. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may occur to those skilled in the art without departing from the principle of the invention, and are considered to be within the scope of the invention.
Claims (10)
1. A multi-level congestion control method in a high-speed interconnection network is characterized in that a source node in the high-speed interconnection network processes a CNP message and comprises the following steps:
s1) receiving the CNP message returned by the destination node;
s2) reading the value of the congestion mark carried in the CNP message, wherein the value of the congestion mark is a numerical value and is obtained by accumulating when each exchange node transmits the message and congestion occurs;
s3) adjusting the message sending rate according to the value of the congestion mark to reduce the message congestion.
2. The method for controlling multi-level congestion in a high-speed interconnection network according to claim 1, wherein the detailed step of step S3) comprises:
s3.1) acquiring the corresponding CCTI stepping length according to the value of the congestion mark;
s3.2) updating the CCTI value according to the CCTI stepping length;
s3.3) controlling the message sending rate according to the CCTI value to reduce message congestion.
3. The method for controlling multi-level congestion in a high-speed interconnection network according to claim 1, wherein step S1) is preceded by the step of sending a CNP message by a destination node in the high-speed interconnection network:
t1) reading the value of the congestion mark in the message header slice of the received message, and skipping to execute the next step when the value of the congestion mark in the message header slice of the message is larger than a preset initial value;
t2) generates a CNP message, and adds the value of the congestion flag in the header slice of the received message to the CNP message, and transmits the value of the congestion flag in the header slice of the received message back to the source node through the CNP message.
4. The method according to claim 3, wherein the step T2) of attaching the value of the congestion flag in the header slice of the received packet to the CNP packet means attaching the value of the congestion flag in the header slice of the received packet to a multi-bit field specified in the header slice of the CNP packet.
5. The method for controlling multi-level congestion in a high-speed interconnection network according to claim 3, wherein the step of forwarding the packet by the switching node in the high-speed interconnection network before the step T1) comprises:
E1) receiving a message from the previous hop;
E2) judging whether the congestion condition occurs at present, and if the congestion condition does not occur, skipping to execute the step E3); otherwise, accumulating the congestion mark value in the message header slice of the message;
E3) and outputting the message to the next hop.
6. The method for controlling multi-level congestion in a high-speed interconnection network according to claim 5, wherein the detailed step of accumulating the values of the congestion flags in the slice header of the packet in step E2) comprises: reading the value of the congestion mark in the message header slice of the message, judging whether the value of the congestion mark is equal to the maximum value which can be stored by the congestion mark, if the value of the congestion mark is equal to the maximum value which can be stored by the congestion mark, keeping the value of the congestion mark in the message header slice of the message unchanged, and skipping to execute the step E3); otherwise, adding 1 to the value of the congestion mark on the basis of the original value, updating the congestion mark in the message header slice of the message, and skipping to execute the step E3).
7. A multi-level congestion control apparatus in a high speed interconnect network, comprising:
a CNP message receiving program unit, configured to receive a CNP message returned by a destination node;
a CNP message analysis program unit, configured to read a value of a congestion flag carried in a CNP message, where the value of the congestion flag is a numerical value and is obtained by accumulating congestion occurring in a transmission message through each switching node;
and the message rate control program unit is used for adjusting the rate of sending the message according to the value of the congestion mark so as to reduce the message congestion.
8. A multi-level congestion control apparatus in a high speed interconnection network, wherein the multi-level congestion control apparatus is programmed or configured to perform the steps of the multi-level congestion control method in the high speed interconnection network according to any one of claims 1 to 6, or a computer program or firmware program programmed or configured to perform the multi-level congestion control method in the high speed interconnection network according to any one of claims 1 to 6 is stored in a memory of the multi-level congestion control apparatus.
9. A high speed interconnect network system comprising a source node, a destination node and a switching node, the source node passing through at least one switching node and a destination node, characterized in that the source node is programmed or configured to perform the steps of the multi-level congestion control method in the high speed interconnect network of claim 1 or 2, or the destination node is programmed or configured to perform the steps of the multi-level congestion control method in the high speed interconnect network of claim 3 or 4, or the switching node is programmed or configured to perform the steps of the multi-level congestion control method in the high speed interconnect network of claim 5 or 6.
10. A computer-readable storage medium having stored thereon a computer program or firmware program programmed or configured to perform the method for multi-level congestion control in a high speed interconnect network according to any of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010456010.1A CN111683017B (en) | 2020-05-26 | 2020-05-26 | Multi-level congestion control method, device, system and medium in high-speed interconnection network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010456010.1A CN111683017B (en) | 2020-05-26 | 2020-05-26 | Multi-level congestion control method, device, system and medium in high-speed interconnection network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111683017A true CN111683017A (en) | 2020-09-18 |
CN111683017B CN111683017B (en) | 2023-04-07 |
Family
ID=72453255
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010456010.1A Active CN111683017B (en) | 2020-05-26 | 2020-05-26 | Multi-level congestion control method, device, system and medium in high-speed interconnection network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111683017B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113114578A (en) * | 2021-03-29 | 2021-07-13 | 紫光华山科技有限公司 | Traffic congestion isolation method, device and system |
CN113938432A (en) * | 2021-12-02 | 2022-01-14 | 中国人民解放军国防科技大学 | High-speed interconnection network congestion control marking method and device |
CN116886615A (en) * | 2023-09-07 | 2023-10-13 | 苏州浪潮智能科技有限公司 | Network congestion control method, device, equipment, system and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101188611A (en) * | 2007-11-21 | 2008-05-28 | 华为技术有限公司 | Congestion notification method, system and node device |
US20080267073A1 (en) * | 2007-04-25 | 2008-10-30 | Patricia Ann Thaler | Method and System for Ethernet Congestion Management |
CN102594713A (en) * | 2012-03-29 | 2012-07-18 | 杭州华三通信技术有限公司 | Method and devices for achieving explicit congestion notification |
CN110505156A (en) * | 2019-08-23 | 2019-11-26 | 中国人民解放军国防科技大学 | Low-overhead congestion control method in high-speed interconnection network |
-
2020
- 2020-05-26 CN CN202010456010.1A patent/CN111683017B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080267073A1 (en) * | 2007-04-25 | 2008-10-30 | Patricia Ann Thaler | Method and System for Ethernet Congestion Management |
CN101188611A (en) * | 2007-11-21 | 2008-05-28 | 华为技术有限公司 | Congestion notification method, system and node device |
CN102594713A (en) * | 2012-03-29 | 2012-07-18 | 杭州华三通信技术有限公司 | Method and devices for achieving explicit congestion notification |
CN110505156A (en) * | 2019-08-23 | 2019-11-26 | 中国人民解放军国防科技大学 | Low-overhead congestion control method in high-speed interconnection network |
Non-Patent Citations (4)
Title |
---|
ZIHAO WEI,ET.AL: "《EC4: ECN and Credit-Reservation Converged Congestion Control》" * |
ZIHAO WEI;ET.AL: "EC4: ECN and Credit-Reservation Converged Congestion Control" * |
刘梦娟等: "《分层组播拥塞控制机制研究》" * |
刘梦娟等: "分层组播拥塞控制机制研究" * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113114578A (en) * | 2021-03-29 | 2021-07-13 | 紫光华山科技有限公司 | Traffic congestion isolation method, device and system |
CN113938432A (en) * | 2021-12-02 | 2022-01-14 | 中国人民解放军国防科技大学 | High-speed interconnection network congestion control marking method and device |
CN113938432B (en) * | 2021-12-02 | 2024-01-02 | 中国人民解放军国防科技大学 | Congestion control marking method and device for high-speed interconnection network |
CN116886615A (en) * | 2023-09-07 | 2023-10-13 | 苏州浪潮智能科技有限公司 | Network congestion control method, device, equipment, system and storage medium |
CN116886615B (en) * | 2023-09-07 | 2024-02-02 | 苏州浪潮智能科技有限公司 | Network congestion control method, device, equipment, system and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN111683017B (en) | 2023-04-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111683017B (en) | Multi-level congestion control method, device, system and medium in high-speed interconnection network | |
US11646967B2 (en) | Packet control method and network apparatus | |
EP2615802B1 (en) | Communication apparatus and method of content router to control traffic transmission rate in content-centric network (CCN), and content router | |
KR101203471B1 (en) | Method of transmitting ethernet frame in a network bridge and the bridge apparatus | |
CN110505156B (en) | Low-overhead congestion control method in high-speed interconnection network | |
CN111800351B (en) | Congestion notification packet generation by a switch | |
US20120147752A1 (en) | Method and apparatus for indicating congestion in a source routed network | |
CN107948103B (en) | Switch PFC control method and control system based on prediction | |
KR20040071220A (en) | Method and apparatus for network congestion control | |
CA2299041A1 (en) | Method and apparatus for forecasting and controlling congestion in a data transport network | |
CN111935031B (en) | NDN architecture-based traffic optimization method and system | |
US20230370387A1 (en) | Network congestion handling method, apparatus, and device | |
CN111026324B (en) | Updating method and device of forwarding table entry | |
US11258723B2 (en) | Data processing method and apparatus, and switching device using footprint queues | |
CN111585911A (en) | Method for balancing network traffic load of data center | |
JP2011035608A (en) | Edge node, window size control method, and program | |
Ferrer et al. | Progressive congestion management based on packet marking and validation techniques | |
CN112073325B (en) | Data congestion control and bandwidth estimation method | |
CN113612698A (en) | Data packet sending method and device | |
CN111726300A (en) | Data sending method and device | |
US20230236727A1 (en) | Circuit, Chip, and Electronic Device | |
CN110086857B (en) | Information center network cache deployment method based on cache node state and position | |
US11706144B2 (en) | Marking packets based on egress rate to indicate congestion | |
CN113938432B (en) | Congestion control marking method and device for high-speed interconnection network | |
US20220353198A1 (en) | Forwarding Information Obtaining Method and Apparatus |
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 |