CN113225241A - Data transmission congestion control method and system for ring-shaped data message network - Google Patents
Data transmission congestion control method and system for ring-shaped data message network Download PDFInfo
- Publication number
- CN113225241A CN113225241A CN202110417613.5A CN202110417613A CN113225241A CN 113225241 A CN113225241 A CN 113225241A CN 202110417613 A CN202110417613 A CN 202110417613A CN 113225241 A CN113225241 A CN 113225241A
- Authority
- CN
- China
- Prior art keywords
- node
- virtual link
- ring
- network
- message
- 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
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/42—Loop networks
-
- 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
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/104—Peer-to-peer [P2P] networks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Small-Scale Networks (AREA)
Abstract
The invention provides a data transmission congestion control method and a data transmission congestion control system for a ring-shaped data message network, which comprise an active congestion control technical scheme, wherein the maximum available bandwidth of a virtual link is actively defined, so that a plurality of virtual links can avoid node congestion while maximally utilizing the link bandwidth; the technical scheme for processing link contention utilizes a counter to realize weighted average occupation of different virtual links in the same node on a transmitting port TX within a period of time, so that link contention is prevented.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a method and a system for controlling data transmission congestion for a ring-shaped data packet network.
Background
In various computer systems, a network card device is generally provided on a computer motherboard in order to communicate with the outside. Typically, network card devices establish communications using a data packet network.
Common modes for networking a plurality of nodes include bus networking, star topology networking and ring topology networking. The bus networking is basically not used in the wired network networking because the conflict problem needs to be solved; the ring topology network can alternately obtain the control right of the ring by a token ring method, but because of low efficiency, the token ring network is hardly used in the current high-speed networking; the star topology network has excellent switching performance and lower cost under the conventional ethernet environment, and thus becomes a mainstream networking mode of a small network. However, when networking a high-speed network, due to high-frequency signal crosstalk caused by electrical characteristics of the circuit, a star topology using the circuit as a transmission medium greatly increases networking cost, and is difficult to miniaturize.
The optical fiber is used as a novel transmission medium, high-frequency crosstalk in a circuit network does not exist naturally, and the optical fiber is an ideal medium for high-speed network networking, but because the physical characteristics of light are difficult to realize a temporary storage unit of an optical medium, the design of an optical star topology switch is difficult, especially the message forwarding of a data message network aiming at optical signal networking has design difficulty, and at the moment, the ring topology network becomes one of networking strategies of an optical interconnection network again due to the advantages of low cost and simple networking. For optical signal networking, if optical star topology networking is adopted, the optical star topology networking must enter a circuit domain, higher requirements on a switch are met, and miniaturization is difficult to achieve; the ring topology solves the problem of miniaturization, but deals with the problem of link contention and improves the quality of service of the network.
The pure ring topology networking makes full use of the buffer area inside each node in the network, and each node participates in both data receiving and transmitting and data forwarding by using a link multiplexing technology. However, this also exacerbates the problem of contention for the same transmit port (TX) for both forwarding traffic and local traffic. Meanwhile, for any node in the ring topology network, the maximum carrying capacity of the transmitting port TX can be easily exceeded by the traffic peak value received at the receiving port (RX) of the node and the locally generated traffic peak value to be transmitted to another node. If no suitable congestion control mechanism exists, a single node is easy to generate congestion to cause buffer overflow, and further packet loss is generated, and the availability of the ring topology network is very poor.
For the link contention problem, the token ring network avoids the problem through an inefficient method: a "token" is set to be passed in a node of the network and provision is made that a data packet can be sent only if the node owns the "token". Obviously, for a node that does not own a token, even though its TX port link load is 0, this portion of the link is wasted because no "token" can not send data. Obviously, this is not a good solution to the dispute.
For the problem of congestion control, a ring topology network does not have a dedicated congestion control mechanism which is fully adaptive to the characteristics of the ring topology network, but an upper layer protocol is adopted to process congestion. In an application scenario of adopting a ring network topology, the networking scale is often small (several to dozens of nodes), the distance between the nodes is often short (meter level or even lower), the nodes are also often highly isomorphic (adopting completely consistent network cards and physical links), and at this time, a general congestion control mechanism designed for adapting to a large-scale heterogeneous network cannot fully exploit the potential of the ring topology network. Taking the congestion control mechanism in the DCQCN protocol as an example, the DCQCN protocol is a passive control protocol, and like the sliding window mechanism of the TCP protocol, there are an acceleration transmission interval and a congestion control interval: when the receiving rate of any node is greater than the sending rate, the node can actively generate a data packet and transmit the data packet back to the previous node to inform the previous node that the congestion is generated, and the previous node is required to be matched with the reduction of the transmission rate; when there is no congestion for a period of time, the previous node will actively increase the transmission rate. When the DCQCN protocol is used for transmission, different flows cannot be effectively communicated with each other, namely, the optimal solution cannot be achieved immediately; while fluctuating repeatedly in the process of achieving an optimal solution. When small-scale data connections are repeatedly established between nodes, such repeated fluctuations make it difficult for the data links to achieve maximum utilization.
Disclosure of Invention
The invention aims to solve the problem of low efficiency caused by congestion of the existing ring topology network, and provides an active congestion control method and system.
Aiming at the defects of the prior art, the invention provides a data transmission congestion control method facing a ring-shaped data message network, which comprises the following steps:
and 4, waiting for the next time period, and executing the step 3 again after the counting value is updated.
The data transmission congestion control method facing the ring-shaped data message network comprises the following steps that when a virtual link is newly added to the ring-shaped data message network:
the source node recalculates the threshold values of all the virtual links in the annular data message network according to the information of the newly added virtual link, generates a control message containing description of the newly added virtual link, continuously transmits the control message among adjacent nodes in the annular data message network, and finally returns to the source node; every time a node passes through, the node updates local global information according to the content of the control message, forwards the control message and updates the threshold values of all the virtual links stored locally.
The data transmission congestion control method facing the ring-shaped data packet network comprises the following steps that for any node m in the ring-shaped data packet network, for all virtual links taking the node m as a starting point and virtual links i passing the node m, the following steps exist:
and Fi is the maximum available bandwidth of the virtual link i, and Fm is the maximum available bandwidth of the physical link of the node m.
It is also specified that for the same virtual link i, the threshold M on any node it passes through is equal,
the data transmission congestion control method facing the annular data message network is characterized in that a node in the annular data message network is a CPU.
The data transmission congestion control method facing the annular data message network is characterized in that nodes form the annular data message network through optical signal networking.
The invention also provides a data transmission congestion control system facing the ring-shaped data message network, which comprises the following steps:
the module 1 is used for acquiring a ring-shaped data message network and constructing a point-to-point virtual link between nodes in the ring-shaped data message network;
a module 2, configured to generate a threshold and a count value corresponding to a virtual link, and determine, every time period, whether the count value is smaller than the threshold, and if so, clear the count value, where the threshold is the maximum data transmission amount of the virtual link in the time period;
a module 3, configured to, when a node in the virtual link wants to send a message, determine whether a count value of the virtual link is smaller than a threshold, if so, send the message, and meanwhile, add the count value to a length value of the message, otherwise, execute a module 4;
and the module 4 is used for waiting for the next time period, and calling the module 3 again after the counting value is updated.
The data transmission congestion control system facing the ring-shaped data message network comprises the following steps that when a virtual link is newly added to the ring-shaped data message network:
the source node recalculates the threshold values of all the virtual links in the annular data message network according to the information of the newly added virtual link, generates a control message containing description of the newly added virtual link, continuously transmits the control message among adjacent nodes in the annular data message network, and finally returns to the source node; every time a node passes through, the node updates local global information according to the content of the control message, forwards the control message and updates the threshold values of all the virtual links stored locally.
The data transmission congestion control system facing the ring-shaped data packet network comprises the following steps that for any node m in the ring-shaped data packet network, for all virtual links taking the node m as a starting point and virtual links i passing through the node m, the following steps are performed:
and Fi is the maximum available bandwidth of the virtual link i, and Fm is the maximum available bandwidth of the physical link of the node m.
It is also specified that for the same virtual link i, the threshold M on any node it passes through is equal,
the data transmission congestion control system facing the ring-shaped data message network is characterized in that a node in the ring-shaped data message network is a CPU.
The data transmission congestion control system facing the annular data message network is characterized in that nodes form the annular data message network through optical signal networking.
According to the scheme, the invention has the advantages that:
after the technical scheme of active congestion control and link contention processing is added to the high-speed high-reliability small-scale ring topology network with highly isomorphic nodes, lower and more stable transmission delay and higher bandwidth utilization rate in the network can be realized, and the performance of the ring topology network is improved to the degree of carrying data exchange in a data center.
Drawings
Fig. 1 is a schematic diagram of a data packet network according to an embodiment of the present invention.
Detailed Description
The poor performance of the conventional ring topology network is caused by the characteristics of the ring topology network itself. The inventor notices that in the field of high-speed networks, optical fiber transmission has unparalleled anti-interference characteristics, but through evaluation, the star topology switch widely used in the field of electrical signal transmission is not suitable for optical signal networking in terms of cost and efficiency due to the physical properties of the optical fiber. However, after sufficient research, the inventor finds that a ring topology networking mode which is almost abandoned in the field of electric signals has own unique advantages in optical signal networking, and simultaneously notes that a communication technical means suitable for the traditional ring topology network does not synchronously evolve along with the evolution of communication infrastructure, and the prior art does not sufficiently utilize information provided inside the network.
The inventor finds out through full research on an Ethernet protocol stack and network topology that the link contention problem of a ring topology network can be solved by utilizing the self characteristics of a high-speed and high-reliability network and the provided additional information, and meanwhile, the active congestion control technology can be realized to radically avoid the packet loss problem. The scheme is as follows:
point-to-point connections between different nodes are defined as virtual links, and a virtual link means that point-to-point connection links between two nodes are virtualized. Since in a ring topology network, a point-to-point physical connection of any two nodes is likely not to exist really. Meanwhile, each node (including a sending node, a forwarding node and a receiving node) participating in the point-to-point connection maintains a section of information related to the virtual link, and then is used for active congestion control, wherein the node refers to a computing node, for example, a node taking a CPU as a core;
adding a register group for recording extra information in an Ethernet link layer, wherein the register group is used for respectively recording necessary information of each global virtual link; wherein the additional information refers to information outside of the definition of the IEEE 802.3 specification with respect to Ethernet; the necessary information is a subset of the relevant information. The additional information emphasizes the concrete hardware implementation, the related information emphasizes the abstract data structure, and the necessary information emphasizes that the information is enough to complete the packet forwarding function between the adjacent nodes.
Each time a new point-to-point connection is established, the corresponding virtual link is established, and all the related nodes in the whole situation update the information of the whole situation link at the same time;
when each node has the global virtual link information, the most suitable available bandwidth of each virtual link can be calculated, and the calculation results of all the nodes are the same. At this time, the sending rate of the virtual link sending end is actively controlled by using the calculation result, so that the sum of the RX flow of any node and the flow generated by the node is ensured to be less than or equal to the maximum processing capacity of the TX port of the node, and the packet loss caused by buffer overflow is further avoided fundamentally.
The invention comprises the following steps: the active congestion control technical scheme actively defines the maximum available bandwidth of the virtual links, so that the congestion of nodes is avoided while the bandwidth of the links is maximally utilized by a plurality of virtual links; the technical scheme for processing link contention utilizes a counter to realize weighted average occupation of different virtual links in the same node on a transmitting port TX within a period of time, so that link contention is prevented.
In order to make the aforementioned features and effects of the present invention more comprehensible, embodiments accompanied with figures are described in detail below.
Before the communication between the nodes is carried out, virtual links between the nodes are firstly established (the virtual links are realized by the nodes internally and are transparent to upper layer protocols). Fig. 1 shows an 8-node ring topology network on which 6 virtual links are constructed, 1 to 4, 1 to 6, 1 to 8, 2 to 3, 2 to 7, and 6 to 8.
Each node has a separate area for storing virtual link information of the current entire network. Each virtual link is stored using a tuple: { source address identifier, destination address identifier, expected bandwidth, actual available bandwidth }. Meanwhile, for a virtual link which is a source node of the node itself or passes through the node itself, the node should independently store an associated counter Cnt and an associated threshold value M for the node.
The expected bandwidth is the maximum bandwidth expected to be obtained by the point-to-point virtual link; the actual available bandwidth is the maximum bandwidth actually obtained by the point-to-point virtual link. Due to the existence of the bandwidth upper limit of the physical link, when the maximum available bandwidth of the physical link cannot meet the sum of the expected bandwidths of the plurality of virtual links carried on the physical link, the maximum available bandwidth of the physical link can be divided equally to the virtual links according to the weighting of the M value. For any virtual link, the allocated bandwidth is the actual available bandwidth. That is, the value of M determines the actual available bandwidth; the expected bandwidth is equal to the actual available bandwidth at low load and less than the actual available bandwidth at high load.
The process of adding the virtual link between the nodes is as follows:
1. the virtual link control information is transmitted between the nodes through a private control protocol. When the virtual link is established for the first time, the source node calculates M values of all the virtual links according to the information of the newly added virtual link. The calculation method is described below. And then, sending a control message to the destination node, wherein the message contains a tuple describing the newly added virtual link.
2. Every time a node passes through, the passing node updates local global information according to the message content, forwards the control message and updates the M values of all the virtual links stored locally. Note that each node will have the latest complete global information at this time, and therefore, using the same update algorithm as the source node, the same calculation result as the source node will be obtained. For a ring topology network, control packets are continuously transmitted (clockwise or counterclockwise) between adjacent nodes. When the control message reaches one of the nodes, the node has the latest updated information and is combined with the currently stored global information to form the latest complete global information. Each control message is transmitted from a node clockwise or counterclockwise, and finally returns to the node after passing through all other nodes.
3. After the control message is finally transmitted back to the source node, the process of newly building the virtual link is completely finished, and the source node does not need to update the locally stored global information.
And at this point, the virtual link from the source node to the destination node is built.
When a virtual link needs to change the expected bandwidth, the process is very similar to that of the newly added virtual link, and only the information of the newly added virtual link transmitted in the control message is replaced by the information of the virtual link with the changed expected bandwidth. The expected bandwidth is determined according to the priority of the tasks running in the node to the transmission. High priority transmissions tend to mean lower delay tolerance, and larger total data transmission volume. If the node can expect the load which can be generated by the node in a future period of time and finds that the expected bandwidth is not matched with the expected bandwidth of the currently established virtual link, the expected bandwidth information needs to be updated, so that the globally available bandwidth can be released in time, or the service quality of the virtual link of the node can be improved.
When a certain virtual link has no data load within a period of time, the source node initiates a destroy request of the virtual link, and the destroy request is also forwarded by the nodes in sequence, and the destroy signal is transmitted to all the other nodes so as to update the M values of the other virtual links.
The congestion control scheme of the invention will be detailed next: the flow of each virtual link is determined by a separate threshold value M, and the value of M may be different for different virtual links. And setting a fixed time interval T seconds for all the virtual links, subtracting M from the value of the counter Cnt corresponding to the virtual link every T seconds, and clearing the Cnt if the value of the Cnt is less than M. The larger the T is, the lower the power consumption is, but the higher the requirement on the size of the buffer area is, the less ideal the time delay is, and the less good the processing effect on the link contention; the smaller the T, the higher the corresponding frequency, the higher the power consumption, but the requirement on the size of the buffer area can be reduced, meanwhile, the data packet forwarding time delay can be reduced, and the processing effect on the link contention is improved. It is generally recommended to set T to be the time required to send a packet, or to be of an order of magnitude similar thereto.
When a data message belonging to the virtual link is received from RX each time, if the corresponding Cnt value is smaller than M, forwarding the message, and simultaneously increasing the length value of the message by Cnt; if the value of Cnt is larger than or equal to M, the message is not allowed to be forwarded at this time, and the message is placed into a buffer zone. And judging whether the virtual link is allowed to be forwarded again or not again after the counter Cnt is updated until the next time T.
And when the node is the source node of the virtual link, the same judgment logic is adopted. The only difference is that when forwarding is not allowed, the operation of placing the packet into a buffer is omitted, since the data is already present in the buffer of the source node.
If the maximum available bandwidth of the physical link is Fm (bits/s), the physical link can transmit Fm × T bits at most in T time. And setting the maximum available bandwidth of a certain virtual link i as Fi (bits/s), and if Fi < Fm, the virtual link i can transmit Fi T bits at most in T time. Let M be Fi × T, the forwarding rule may limit the actual transmission bandwidth of the virtual link i to Fi bits/s on a macroscopic time scale. Rules are additionally formulated so that any given node m exists for all virtual links starting at m and for all virtual links i passing through m
Fi refers to the maximum available bandwidth that the virtual link is allowed to reach after the link is constructed. Fm pointer is to a point-to-point physical link, which can reach the maximum available bandwidth. A virtual link may span several point-to-point physical links. For a point-to-point physical link connecting nodes m and n, data can only go from m to n. At this time, on the physical link, there are necessarily only two types of virtual links: a virtual link that starts at m, and a virtual link that does not start at m but passes through an m-node.
And if M is equal at any node passing through the same virtual link i, the ring topology network has the following properties:
any virtual link will not generate backlog at any node that the link passes through, causing its buffer to overflow;
the sum of the bandwidth of the receiving port RX of any node and the bandwidth of the traffic generated by itself does not exceed the maximum available bandwidth of the transmitting port TX.
At this time, no buffer backlog is generated at any position of the network, so that packet loss caused by processing capacity limitation does not exist in the network, that is, active congestion control is realized.
The following are system examples corresponding to the above method examples, and this embodiment can be implemented in cooperation with the above embodiments. The related technical details mentioned in the above embodiments are still valid in this embodiment, and are not described herein again in order to reduce repetition. Accordingly, the related-art details mentioned in the present embodiment can also be applied to the above-described embodiments.
The invention also provides a data transmission congestion control system facing the ring-shaped data message network, which comprises the following steps:
the module 1 is used for acquiring a ring-shaped data message network and constructing a point-to-point virtual link between nodes in the ring-shaped data message network;
a module 2, configured to generate a threshold and a count value corresponding to a virtual link, and determine, every time period, whether the count value is smaller than the threshold, and if so, clear the count value, where the threshold is the maximum data transmission amount of the virtual link in the time period;
a module 3, configured to, when a node in the virtual link wants to send a message, determine whether a count value of the virtual link is smaller than a threshold, if so, send the message, and meanwhile, add the count value to a length value of the message, otherwise, execute a module 4;
and the module 4 is used for waiting for the next time period, and calling the module 3 again after the counting value is updated.
The data transmission congestion control system facing the ring-shaped data message network comprises the following steps that when a virtual link is newly added to the ring-shaped data message network:
the source node recalculates the threshold values of all the virtual links in the annular data message network according to the information of the newly added virtual link, generates a control message containing description of the newly added virtual link, continuously transmits the control message among adjacent nodes in the annular data message network, and finally returns to the source node; every time a node passes through, the node updates local global information according to the content of the control message, forwards the control message and updates the threshold values of all the virtual links stored locally.
The data transmission congestion control system facing the ring-shaped data packet network comprises the following steps that for any node m in the ring-shaped data packet network, for all virtual links taking the node m as a starting point and virtual links i passing through the node m, the following steps are performed:
and Fi is the maximum available bandwidth of the virtual link i, and Fm is the maximum available bandwidth of the physical link of the node m.
And the M value on any node passing by the same virtual link i is equal,
the data transmission congestion control system facing the ring-shaped data message network is characterized in that a node in the ring-shaped data message network is a CPU.
The data transmission congestion control system facing the annular data message network is characterized in that nodes form the annular data message network through optical signal networking.
Claims (10)
1. A data transmission congestion control method facing a ring-shaped data message network is characterized by comprising the following steps:
step 1, acquiring the annular data message network, and constructing a point-to-point virtual link between nodes in the annular data message network;
step 2, the virtual link has a corresponding threshold value and a count value, and whether the count value is smaller than the threshold value is judged every time period, if yes, the count value is cleared, wherein the threshold value is the maximum data transmission quantity of the virtual link in the time period;
step 3, when the node in the virtual link wants to send a message, judging whether the count value of the virtual link is smaller than a threshold value, if so, sending the message, and simultaneously increasing the length value of the message by the count value, otherwise, executing the step 4;
and 4, waiting for the next time period, and executing the step 3 again after the counting value is updated.
2. The method for controlling data transmission congestion of ring-shaped data message network as claimed in claim 1, wherein when a virtual link is added to the ring-shaped data message network:
the source node recalculates the threshold values of all the virtual links in the annular data message network according to the information of the newly added virtual link, generates a control message containing description of the newly added virtual link, continuously transmits the control message among adjacent nodes in the annular data message network, and finally returns to the source node; every time a node passes through, the node updates local global information according to the content of the control message, forwards the control message and updates the threshold values of all the virtual links stored locally.
3. The method for controlling data transmission congestion of a ring-shaped data message network according to claim 1, wherein for any node m in the ring-shaped data message network, for all virtual links starting from the node m and virtual links i passing through the node m, there are:
and Fi is the maximum available bandwidth of the virtual link i, and Fm is the maximum available bandwidth of the physical link of the node m.
It is also specified that the threshold values on any node through which it passes are equal for the same virtual link i.
4. The method according to claim 1, wherein the node in the ring-shaped data packet network is a CPU.
5. The method according to claim 1, wherein the nodes form the ring-shaped data packet network by optical signal networking.
6. A data transmission congestion control system for a ring-shaped data packet network, comprising:
the module 1 is used for acquiring a ring-shaped data message network and constructing a point-to-point virtual link between nodes in the ring-shaped data message network;
a module 2, configured to generate a threshold and a count value corresponding to a virtual link, and determine, every time period, whether the count value is smaller than the threshold, and if so, clear the count value, where the threshold is the maximum data transmission amount of the virtual link in the time period;
a module 3, configured to, when a node in the virtual link wants to send a message, determine whether a count value of the virtual link is smaller than a threshold, if so, send the message, and meanwhile, add the count value to a length value of the message, otherwise, execute a module 4;
and a module 4, configured to wait for a next time period, and call the module 3 again after the count value is updated.
7. The data transmission congestion control system of claim 7, wherein when the ring data packet network adds a new virtual link:
the source node recalculates the threshold values of all the virtual links in the annular data message network according to the information of the newly added virtual link, generates a control message containing description of the newly added virtual link, continuously transmits the control message among adjacent nodes in the annular data message network, and finally returns to the source node; every time a node passes through, the node updates local global information according to the content of the control message, forwards the control message and updates the threshold values of all the virtual links stored locally.
8. The data transmission congestion control system of claim 6, wherein for any node m in the ring-shaped data packet network, there are:
and Fi is the maximum available bandwidth of the virtual link i, and Fm is the maximum available bandwidth of the physical link of the node m.
It is also specified that the threshold values on any node through which it passes are equal for the same virtual link i.
9. The data transmission congestion control system of claim 6, wherein the nodes in the ring data packet network are CPUs.
10. The data transmission congestion control system of claim 6, wherein the nodes form the ring-shaped data message network by networking through optical signals.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110417613.5A CN113225241B (en) | 2021-04-19 | 2021-04-19 | Data transmission congestion control method and system for ring-shaped data message network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110417613.5A CN113225241B (en) | 2021-04-19 | 2021-04-19 | Data transmission congestion control method and system for ring-shaped data message network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113225241A true CN113225241A (en) | 2021-08-06 |
| CN113225241B CN113225241B (en) | 2022-09-06 |
Family
ID=77087861
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110417613.5A Active CN113225241B (en) | 2021-04-19 | 2021-04-19 | Data transmission congestion control method and system for ring-shaped data message network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113225241B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116319158A (en) * | 2023-05-23 | 2023-06-23 | 南京芯驰半导体科技有限公司 | Node data processing method, device, equipment and medium based on daisy chain network |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6650618B1 (en) * | 1999-07-01 | 2003-11-18 | Nortel Networks Limited | Method and apparatus for managing communications between nodes in a bi-directional ring network |
| CN1496067A (en) * | 2002-08-02 | 2004-05-12 | ���Ͽع�����˾ | System and method for increasing block ring network fairness |
| US20080298231A1 (en) * | 2007-04-24 | 2008-12-04 | Fujitsu Limited | Ring node and redundancy method |
| CN101605102A (en) * | 2009-07-16 | 2009-12-16 | 杭州华三通信技术有限公司 | Load sharing method during a kind of IRF piles up and device |
| US20100309776A1 (en) * | 2009-06-05 | 2010-12-09 | Fahd Alharbi | Allocating Bandwidth in a Resilient Packet Ring Network by P Controller |
| CN102035741A (en) * | 2010-12-20 | 2011-04-27 | 中兴通讯股份有限公司 | Unicast message forwarding method and device in ring topology network |
| CN102857436A (en) * | 2012-09-10 | 2013-01-02 | 杭州华三通信技术有限公司 | Flow transmission method and flow transmission equipment based on IRF (intelligent resilient framework) network |
| CN103428103A (en) * | 2013-08-08 | 2013-12-04 | 杭州华三通信技术有限公司 | Link load control method and stacking device |
| CN104243334A (en) * | 2014-09-26 | 2014-12-24 | 杭州华三通信技术有限公司 | Method and equipment for preventing link congestion |
| WO2015196647A1 (en) * | 2014-06-24 | 2015-12-30 | 中兴通讯股份有限公司 | Path selection method and apparatus for ring topology stacking system and master device |
| CN106302161A (en) * | 2016-08-01 | 2017-01-04 | 广东工业大学 | Perception data transmission method based on load estimation, device, path control deivce |
| CN106375222A (en) * | 2015-07-20 | 2017-02-01 | 中兴通讯股份有限公司 | Multicast line switching method for annular networking and equipment |
| CN107196854A (en) * | 2017-06-20 | 2017-09-22 | 西安交通大学 | Datum plane abnormality eliminating method in a kind of software defined network |
| WO2018195728A1 (en) * | 2017-04-24 | 2018-11-01 | 华为技术有限公司 | Client service transmission method and device |
| CN109547095A (en) * | 2018-12-06 | 2019-03-29 | 长沙天仪空间科技研究院有限公司 | A kind of method that congestion is alleviated in satellite communication in the process |
-
2021
- 2021-04-19 CN CN202110417613.5A patent/CN113225241B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6650618B1 (en) * | 1999-07-01 | 2003-11-18 | Nortel Networks Limited | Method and apparatus for managing communications between nodes in a bi-directional ring network |
| CN1496067A (en) * | 2002-08-02 | 2004-05-12 | ���Ͽع�����˾ | System and method for increasing block ring network fairness |
| US20080298231A1 (en) * | 2007-04-24 | 2008-12-04 | Fujitsu Limited | Ring node and redundancy method |
| US20100309776A1 (en) * | 2009-06-05 | 2010-12-09 | Fahd Alharbi | Allocating Bandwidth in a Resilient Packet Ring Network by P Controller |
| CN101605102A (en) * | 2009-07-16 | 2009-12-16 | 杭州华三通信技术有限公司 | Load sharing method during a kind of IRF piles up and device |
| CN102035741A (en) * | 2010-12-20 | 2011-04-27 | 中兴通讯股份有限公司 | Unicast message forwarding method and device in ring topology network |
| CN102857436A (en) * | 2012-09-10 | 2013-01-02 | 杭州华三通信技术有限公司 | Flow transmission method and flow transmission equipment based on IRF (intelligent resilient framework) network |
| CN103428103A (en) * | 2013-08-08 | 2013-12-04 | 杭州华三通信技术有限公司 | Link load control method and stacking device |
| CN105323170A (en) * | 2014-06-24 | 2016-02-10 | 中兴通讯股份有限公司 | Path selecting method and apparatus of ring topology stacking system, and master apparatus of ring topology stacking system |
| WO2015196647A1 (en) * | 2014-06-24 | 2015-12-30 | 中兴通讯股份有限公司 | Path selection method and apparatus for ring topology stacking system and master device |
| CN104243334A (en) * | 2014-09-26 | 2014-12-24 | 杭州华三通信技术有限公司 | Method and equipment for preventing link congestion |
| CN106375222A (en) * | 2015-07-20 | 2017-02-01 | 中兴通讯股份有限公司 | Multicast line switching method for annular networking and equipment |
| CN106302161A (en) * | 2016-08-01 | 2017-01-04 | 广东工业大学 | Perception data transmission method based on load estimation, device, path control deivce |
| WO2018195728A1 (en) * | 2017-04-24 | 2018-11-01 | 华为技术有限公司 | Client service transmission method and device |
| CN109314673A (en) * | 2017-04-24 | 2019-02-05 | 华为技术有限公司 | A kind of client traffic transmission method and device |
| CN107196854A (en) * | 2017-06-20 | 2017-09-22 | 西安交通大学 | Datum plane abnormality eliminating method in a kind of software defined network |
| CN109547095A (en) * | 2018-12-06 | 2019-03-29 | 长沙天仪空间科技研究院有限公司 | A kind of method that congestion is alleviated in satellite communication in the process |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116319158A (en) * | 2023-05-23 | 2023-06-23 | 南京芯驰半导体科技有限公司 | Node data processing method, device, equipment and medium based on daisy chain network |
| CN116319158B (en) * | 2023-05-23 | 2023-08-29 | 南京芯驰半导体科技有限公司 | Node data processing method, device, equipment and medium based on daisy chain network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113225241B (en) | 2022-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10334059B2 (en) | Network for transporting ethernet and time sensitive data | |
| CN100452750C (en) | Token transfer and manage method for Ad Hoc network | |
| JP5156332B2 (en) | Packet transfer device | |
| CN102075437B (en) | Communication method, gateway and network | |
| JP6551547B2 (en) | Traffic-aware group reconfiguration in multi-group P2P networks | |
| WO2018233425A1 (en) | Network congestion processing method, device, and system | |
| TW201138364A (en) | Method and system for management based end-to-end sleep limitation in an energy efficient ethernet network | |
| CN103124420A (en) | Wireless on-chip network structuring method | |
| CN113225241B (en) | Data transmission congestion control method and system for ring-shaped data message network | |
| CN113726681B (en) | Network congestion control method and network equipment | |
| CN111464437A (en) | Multipath transmission path optimization method based on forward time delay in vehicle-mounted heterogeneous network | |
| CN105812063B (en) | Network on mating plate system based on statistic multiplexing and communication means | |
| Zhu et al. | Efficient hybrid multicast approach in wireless data center network | |
| CN103532865A (en) | Congestion control method based on socially aware in delay tolerant network | |
| Cai et al. | Survey on converged data center networks with DCB and FCoE: standards and protocols | |
| CN112689275B (en) | Novel non-uniform power forming method for BLE mesh network | |
| Nithin et al. | Efficient load balancing for multicast traffic in data center networks using SDN | |
| CN109068394B (en) | Channel access method based on queue length and collision risk | |
| Guo et al. | Experimental Demonstration of SDN-enabled Reconfigurable Disaggregated Data Center Infrastructure | |
| CN101296189A (en) | Distributed stream processing network appliance and packet transmission method thereof | |
| Shu et al. | Optimal many-to-many personalized concurrent communication in RapidIO-based fat-trees | |
| WO2019080744A1 (en) | Control channel implementation method, apparatus, device, storage medium, and processing method | |
| CN115297065B (en) | Processing device communication interconnection method, processing device communication interconnection device, computer device and storage medium | |
| CN214480674U (en) | Bridge and combined network | |
| CN117221212B (en) | Optical network on chip low congestion routing method and related equipment |
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 |