Distributed chip backboard flow load balancing method and device
Through the control cell communication between FAP and FE and the routing table update, the number of links can be dynamically adjusted, solving the problem of traffic imbalance caused by damage to the backplane connector of distributed equipment, realizing load balancing between inlet and outlet, avoiding Exit congestion.
Patent Information
- Application Number
- CN202411040863.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-07-31
AI Technical Summary
The backplane connectors of existing distributed devices are damaged as the equipment is unplugged and aged, resulting in unbalanced links between the inlet and outlet disks, resulting in unbalanced flow, and in extreme cases, it may lead to outlet congestion.
Through periodic control of cell communication between service chip FAP and cross chip FE, update the routing table, dynamically adjust the number of links that can reach, use realtime link state to calculate realtime score, mask the low score link, and achieve load balancing.
It effectively solves the problem of uneven flow of inlet and outlet disks, ensures that the flow carried by each link is completely equal, and avoids export congestion.
Smart Images

Figure CN119052175A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of communication load balancing, and in particular to a distributed chip backplane traffic load balancing method and device. Background Art
[0002] Distributed devices are widely used in existing transmission networks. The service disk responsible for service forwarding and the cross disk responsible for traffic forwarding are connected through backplane connectors. The backplane high-speed connector will inevitably be damaged and unusable due to the plugging and unplugging of equipment and aging. At this time, the service disks interconnected at both ends are connected, and the inlet and outlet links of the cross disk are unbalanced, resulting in uneven traffic on the backplane. In extreme cases, it will cause outlet congestion.
[0003] In summary, a distributed chip backplane traffic load balancing solution is urgently needed. Summary of the invention
[0004] In response to the above problems, the present invention provides a distributed chip backplane traffic load balancing method and device, which are used to solve the imbalance problem of inlet and outlet disks caused by physical damage to the backplane connector of the distributed device board, the imbalance problem of backplane traffic and the outlet congestion problem caused by the imbalance.
[0005] In a first aspect, a distributed chip backplane traffic load balancing method is provided, the method comprising:
[0006] The service chip FAP is connected through the cross-connect chip FE;
[0007] FAP and FE periodically send control cells to adjacent chips and update the routing table by recording the link of the received control cells and the content of the control cells; link indicates the connection between FAP and FE;
[0008] FE dynamically adjusts the number of reachable links based on the real-time link number of each FAP in the routing table.
[0009] Furthermore, the routing table is located in the FAP chip and the FE chip, and is used to record the connection link and the reachable FAP;
[0010] The routing table uses FAP_ID as the key and stores a bitmap. Each bit in the bitmap represents a link on the chip.
[0011] Furthermore, when the FAP comes online for the first time, the FE allocates the total number of initial reachable links to it by sending a control cell, and assigns the same initial score to each link;
[0012] The total number of initially reachable links includes: when the chip goes online for the first time, it sends a control cell containing the chip ID to each link; when the FAP sends the control cell, the FE considers the link reachable after receiving the control cell; and finally, the total number of initially reachable links to the FAP is obtained.
[0013] Furthermore, FE dynamically adjusts the number of reachable links according to the real-time link number of each FAP in the routing table, including:
[0014] FE detects the link status in real time and calculates the current number of reachable links in real time;
[0015] The number of real-time reachable links includes: the chip sends a control cell containing the chip ID to each link in real time, and the control cell sent by the FAP. If the FE receives the control cell, it considers that the link is reachable to the FAP. If not, it considers that the link is unreachable to the FAP. Finally, the number of real-time reachable links (active link) is obtained;
[0016] FE calculates the realtime score based on the number of real-time reachable links, where realtime score = number of real-time reachable links / total number of initial reachable links*total score;
[0017] The total score is a static configuration parameter of FE, which is set when the device is powered on for the first time;
[0018] When forwarding packets, FE removes the links with link score greater than realtime score from the routing table, thus achieving load balancing.
[0019] The link score is set as follows:
[0020] The step value is calculated by the number of parallel link connections and the total score on FE;
[0021] Step value = total score / number of connections;
[0022] Select a random order and set the link score of each link connection on FE to 0, one step value, two steps value, etc. until all link connections are assigned values.
[0023] Furthermore, the FE dynamically adjusts the number of reachable links according to the real-time link number of each FAP in the routing table, and also includes:
[0024] When a link of the target FAP is unreachable, the FE notifies all FAPs by sending control information cells, and proportionally reduces the number of reachable links from all source FAPs to the target FAP.
[0025] In a second aspect, a distributed chip backplane adopts the above-mentioned distributed chip backplane traffic load balancing method.
[0026] In a third aspect, a distributed chip communication method adopts the above-mentioned distributed chip backplane traffic load balancing method.
[0027] In a fourth aspect, a distributed chip backplane traffic load balancing device includes: a service chip FAP and a cross chip FE; the service chip FAP is connected via the cross chip FE;
[0028] FAP and FE periodically send control cells to adjacent chips and update the routing table by recording the link of the received control cells and the content of the control cells;
[0029] FE dynamically adjusts the number of reachable links based on the real-time link number of each FAP in the routing table.
[0030] In a fifth aspect, an electronic device includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;
[0031] a memory storing a computer program;
[0032] The processor is used to implement the above-mentioned distributed chip backplane traffic load balancing method when executing the computer program stored in the memory.
[0033] In a sixth aspect, a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned distributed chip backplane traffic load balancing method is implemented.
[0034] The present disclosure has at least the following beneficial effects:
[0035] The present disclosure is aimed at distributed equipment and can effectively solve the problem of traffic imbalance caused by inconsistent links on the backplane side of the inlet and outlet disks. For the cross disk, the traffic carried by each link corresponding to the inlet and outlet disks can be completely equal, the traffic is more balanced, and the outlet congestion problem is effectively solved.
[0036] Other features and advantages of the present disclosure will be described in the following description, and partly become apparent from the description, or be understood by implementing the present disclosure. The purpose and other advantages of the present disclosure can be realized and obtained by the structures indicated in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0038] Figure 1 This is a flow chart of the load balancing method according to an embodiment of the present disclosure;
[0039] Figure 2 Design a schematic diagram for the control cell communication mechanism and routing table between distributed equipment FE and FAP;
[0040] Figure 3 This is a schematic diagram of the actual effect of the routing table generated between the distributed equipment FE and FAP;
[0041] Figure 4 This is a schematic diagram of the actual connection effect during initialization between the distributed equipment FE and FAP;
[0042] Figure 5 Data diagram for calculating realtime score for distributed device FE;
[0043] Figure 6 This is a schematic diagram of the effect of achieving load balancing between distributed devices FE and FAP;
[0044] Figure 7 A schematic diagram of the electronic device structure. DETAILED DESCRIPTION
[0045] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.
[0046] The load balancing method provided by the present disclosure can be widely used in products such as OTN and PTN.
[0047] like Figure 1 As shown, a distributed chip backplane traffic load balancing method, the method comprising:
[0048] The service chip FAP is connected through the cross-connect chip FE;
[0049] S101, FAP and FE periodically send control cells to adjacent chips, and update the routing table by recording the connection link of the received control cell and the content of the control cell;
[0050] S102, FE dynamically adjusts the number of reachable links according to the real-time number of links of each FAP in the routing table.
[0051] The specific implementation is as follows:
[0052] A communication mechanism is established at the chip level, including the service chip (hereinafter collectively referred to as FAP) and the cross-connect chip (hereinafter collectively referred to as FE). FAP and FE periodically send control cells to adjacent chips, and automatically calculate and update the routing table by recording the link of the received control cell and the content of the cell.
[0053] FE dynamically adjusts the reachable message according to the actual number of links of each FAP, thereby adjusting the number of reachable links. When a link to the target FAP fails and becomes unreachable, FE notifies all FAPs through the control cell mechanism, and proportionally reduces the number of reachable links from all source FAPs to the target FAP.
[0054] like Figure 2 As shown, FAP and FE periodically send control cells to adjacent chips, and automatically calculate and update the routing table by recording the link of the received control cell and the content of the cell. The routing table here is used in the FAP chip and the FE chip to record which links can be used to reach which FAPs. The routing table usually uses FAP_ID as the key and stores a bitmap. Each bit in the bitmap represents a link on this chip.
[0055] The final generated routing table is as follows Figure 3 As shown, FAP_ID1 can be reached through link1 and link2, FAP_ID2 can be reached through link3 and link4, FAP_ID3 can be reached through link1, link2, link3, and link4, and FAP_ID4 is unreachable.
[0056] like Figure 4 As shown, in order to distinguish the inlet and outlet, FIP0 and FOP0 are the inlet input and output of FAP0 respectively:
[0057] FIP stands for FABRIC INPUT PLANE;
[0058] FOP stands for FABRIC OUTPUT PLANE;
[0059] As can be seen from the figure, the three wires 0, 1, and 2 are connected to FAP0, and the three wires 3, 4, and 5 are connected to FAP1.
[0060] When the FAP goes online for the first time, the FE will allocate the total number of reachable links to it through the control cell mechanism and assign a score to each link on average. Each FAP has a total score of 120 points (initial configuration), and there are 3 links between each FAP and the FE. Therefore, the score of each link is as follows: link0: 40, link1: 80, link2: 120 (FAP0), link3: 40, link41: 80, link5: 120 (FAP1).
[0061] Specifically, when the chip goes online for the first time, it will send a control cell containing the chip ID to each link. When the FAP sends the control cell, the FE will consider the link to be reachable, and finally get a total number of links to the FAP. The reason for allocating scores is related to the subsequent total score. Each link has its own score, and the subsequent load balancing calculation is based on the score.
[0062] FE detects link status in real time and calculates the current number of reachable links (active links) in real time. FE calculates a realtime score based on the number of active links. Realtime score = actual number of links / total number of links*total score. When routing and forwarding, FE will mask the links with link score > realtime score from the routing table to achieve load balancing.
[0063] Specifically, the number of reachable links (active links) is determined by the chip sending a control cell containing the chip ID to each link in real time. When the FAP sends the control cell and the FE receives the control cell, it considers that the link is reachable to the FAP. If it does not receive the control cell, it considers that the link is unreachable to the FAP.
[0064] The actual link number is the number of active links mentioned above, which is updated in real time. The total link is the total number of links calculated when the device is first online. The total score is a static configuration, a fixed value of 120 written by the software when the device is first powered on.
[0065] The link score is the score when the service goes online for the first time. For example, if the total score is 120 and there are 3 links in total, the stepping is 40. There are 3 links between each FAP and FE, and the stepping is 40. Therefore, the score of each link is as follows: link0: 40, link1: 80, link2: 120.
[0066] like Figure 5 As shown in the figure, assuming that one link is broken at the egress, the actual active link is 2. The real time sore is 80 points calculated by the table. The actual effect of routing forwarding is shown in the figure below. Figure 6 As shown in the figure, if link3 is broken, the traffic of FAP1 at the entrance will only flow in from link0 and link1 during actual routing forwarding, so that the inlet and outlet loads are balanced.
[0067] Specifically, the number of reachable links from all source FAPs to the target FAP is reduced proportionally, as shown in the following example:
[0068] Assume that FE and two FAPs are interconnected with three links, and the link distribution is as follows:
[0069] FAP1:link0:40,link1:80,link2:120;
[0070] FAP2:link3:40,link4:80,link5:120;
[0071] If link 2 of FAP1 is disconnected, realtimescore = 2 / 3*120 = 80 points. When calculating the route, FE will mask the reachable links to FAP2 that are greater than 80. Therefore, FAP2 will also synchronize one less link.
[0072] Assuming that there is another FAP3 initially connected to FAP3 with 6 links, the link scores are as follows:
[0073] FAP3:link3:20,link4:40,link5:60;
[0074] link6:80,link7:100,link8:120;
[0075] When FE calculates the route to FAP3, FAP3 will simultaneously reduce two links.
[0076] A distributed chip backplane adopts the above-mentioned distributed chip backplane traffic load balancing method.
[0077] A distributed chip communication method adopts the above-mentioned distributed chip backplane traffic load balancing method.
[0078] like Figure 2 As shown, a distributed chip backplane traffic load balancing device includes: a service chip FAP and a cross chip FE; the service chip FAP is connected through the cross chip FE;
[0079] FAP and FE periodically send control cells to adjacent chips and update the routing table by recording the link of the received control cells and the content of the control cells;
[0080] FE dynamically adjusts the number of reachable links based on the real-time link number of each FAP in the routing table.
[0081] like Figure 7 As shown, the present disclosure provides an electronic device, including a processor 701, a communication interface 702, a memory 703 and a communication bus 704, wherein the processor 701, the communication interface 702 and the memory 703 communicate with each other through the communication bus 704;
[0082] Memory 703, storing computer programs;
[0083] The processor 701 is configured to implement the above method when executing the computer program stored in the memory 703 .
[0084] The present disclosure provides a computer-readable storage medium storing a computer program, wherein the computer program implements the above method when executed by a processor.
[0085] The computer-readable storage medium may be included in the device / apparatus described in the above embodiment; or it may exist independently without being assembled into the device / apparatus. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.
[0086] According to an embodiment of the present disclosure, a computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.
[0087] Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure.
Claims
1. A distributed chip backplane traffic load balancing method, characterized in that: The method comprises: The service chip FAP is connected through the cross-connect chip FE; FAP and FE periodically send control cells to adjacent chips and update the routing table by recording the link of the received control cells and the content of the control cells; link indicates the connection between FAP and FE; FE dynamically adjusts the number of reachable links based on the real-time link number of each FAP in the routing table.
2. A distributed chip backplane traffic load balancing method according to claim 1, characterized in that: The routing table is located in the FAP chip and the FE chip and is used to record the connection link and reachable FAPs; The routing table uses FAP_ID as the key and stores a bitmap. Each bit in the bitmap represents a link on the chip.
3. A distributed chip backplane traffic load balancing method according to claim 1, characterized in that: When the FAP goes online for the first time, the FE allocates the total number of initial reachable links to it by sending control cells and assigns the same initial score to each link; The total number of initially reachable links includes: when the chip goes online for the first time, it sends a control cell containing the chip ID to each link; when the FAP sends the control cell, the FE considers the link reachable after receiving the control cell; and finally, the total number of initially reachable links to the FAP is obtained.
4. A distributed chip backplane traffic load balancing method according to claim 3, characterized in that: FE dynamically adjusts the number of reachable links based on the real-time link number of each FAP in the routing table, including: FE detects link status in real time and calculates the current number of reachable links in real time; The number of real-time reachable links includes: the chip sends a control cell containing the chip ID to each link in real time, and the control cell sent by the FAP. If the FE receives the control cell, it considers that the link is reachable to the FAP. If not, it considers that the link is unreachable to the FAP. Finally, the number of real-time reachable links (active link) is obtained; FE calculates the realtime score based on the number of real-time reachable links, where realtime score = number of real-time reachable links / total number of initial reachable links*total score; The total score is a static configuration parameter of FE, which is set when the device is powered on for the first time; When forwarding packets, FE removes the links with link score greater than realtime score from the routing table, thus achieving load balancing. The link score is set as follows: The step value is calculated by the number of parallel link connections and the total score on FE; Step value = total score / number of connections; Select a random order and set the link score of each link connection on FE to 0, one step value, two steps value, etc. until all link connections are assigned values.
5. A distributed chip backplane traffic load balancing method according to claim 1, characterized in that: FE dynamically adjusts the number of reachable links based on the real-time link number of each FAP in the routing table, and also includes: When a link of the target FAP is unreachable, the FE notifies all FAPs by sending control information cells, and proportionally reduces the number of reachable links from all source FAPs to the target FAP.
6. A distributed chip backplane, characterized in that: A distributed chip backplane traffic load balancing method as described in any one of claims 1 to 5 is adopted.
7. A distributed chip communication method, characterized in that: A distributed chip backplane traffic load balancing method as described in any one of claims 1 to 5 is adopted.
8. A distributed chip backplane traffic load balancing device, characterized in that: include: Business chip FAP and cross chip FE; business chip FAP is connected through cross chip FE; FAP and FE periodically send control cells to adjacent chips and update the routing table by recording the connection link and content of the received control cells. FE dynamically adjusts the number of reachable links based on the real-time link number of each FAP in the routing table.
9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; a memory storing a computer program; The processor is used to implement a distributed chip backplane traffic load balancing method as described in any one of claims 1 to 5 when executing a computer program stored in a memory.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, a distributed chip backplane traffic load balancing method according to any one of claims 1 to 5 is implemented.
Citation Information
Patent Citations
Method of data storing and maintenance in a distributed data storage system and corresponding device
CN104364765A
Physical link flow balancing method and device
CN111654449A
Wireless CPE and method for realizing load balancing based on wireless CPE
CN112383933A
Distributed storage system load balancing scheduling method and device and storage medium
CN114500578A
Data transmission system, method and device
CN115208839A
