Peer-to-peer optical interconnection circuit switching networking method and system based on DPU

Abstract

The invention discloses a DPU-based peer-to-peer optical interconnection circuit switching networking method and system, and the method comprises the steps: collecting the networking information of an optical interconnection circuit switching network, building a peer-to-peer networking model according to the networking information, and carrying out the peer-to-peer networking between a main network and an auxiliary network according to the peer-to-peer networking model. Setting access nodes carrying DPU acceleration cards for interconnection and intercommunication; and the DPU adopts peer-to-peer traffic agreement to realize a peer-to-peer optical interconnection circuit interaction network. According to the method and the system, a network global routing model is locally maintained at each DPU-based access node, and routing query is moved to the local, so that meaningless communication flow is avoided, the pressure of an auxiliary network is reduced, the available reconstruction frequency of the network is improved, and the network switching performance is improved.

Description

Technical Field

[0001] The present application relates to the technical field of circuit switching, and in particular to a DPU-based peer-to-peer optical interconnection circuit switching networking method and system thereof. Background Art

[0002] Currently, as data centers put forward further requirements for internal interconnection bandwidth and interconnection scale, circuit switching networks based on reconfigurable optical interconnection have successfully overcome the problems of bandwidth, power consumption and interconnection scale and become the mainstream technology.

[0003] However, circuit switching networks based on reconfigurable optical interconnection also bring new problems: the working principle of circuit switching networks is different from that of data packet switching networks. The former is equivalent to a set of dynamically created point-to-point links and does not cache the traffic passing through it, while the latter caches and forwards the traffic passing through it, resulting in the natural separation of the control plane and data plane of the exchange under this interconnection architecture. The access node cannot throw all the switching work to the switch like the traditional data packet switching network, but needs to handle a lot of related work itself.

[0004] The main problems of the prior art are as follows:

[0005] 1. Circuit-switched networks usually require an auxiliary network to coordinate and synchronize the global routing status. The network needs to meet the requirements of near-real-time full interconnection and extremely low latency, so it is usually implemented using a datagram switching network (that is, an existing switch network). The arbitration unit of the circuit-switched network is usually designed in the datagram switch to make arbitration decisions on conflicting routes quickly and efficiently. However, the arbitration unit independently set in the switch will cause the access node to generate a large amount of meaningless routing query traffic, causing communication pressure on the auxiliary network and reducing the global routing efficiency;

[0006] 2. In the previous network structure, the access node does not care about the global routing status of the network but only cares about its own current routing endpoint. When it is necessary to change its own routing endpoint, it will ask the arbitration unit on the switch side (usually implemented by a dedicated ASIC chip or FPGA chip, or by a CPU or DPU) whether it can be changed. Once the request is rejected, it means that the request and response belong to meaningless query traffic. A large amount of meaningless query traffic will cause the auxiliary network to be overloaded, thereby significantly increasing the communication delay of the auxiliary network, and making the current routing status of the optical interconnection network unable to be synchronized to the access node side in time. Once the routing status on the node side is out of sync, it will cause data to be sent to the wrong destination or even data corruption. The previous circuit switching network therefore reserved a longer synchronization time window to avoid accidental communications without synchronization, but accordingly, when switching the routing endpoint, the communication delay of the optical network will increase significantly.

[0007] In response to the above two problems, there is an urgent need for a new type of hardware DPU configured at the access node end, which will be able to undertake corresponding tasks, reduce meaningless traffic in the auxiliary network, improve the communication efficiency of the auxiliary network, shorten the synchronization time window, and thus improve the global routing efficiency. Method and system. Summary of the invention

[0008] In order to solve the problem in the prior art that the routing synchronization cost in the circuit switching network increases significantly with the increase of the interconnection scale, the present invention proposes a DPU-based peer-to-peer circuit switching network networking method.

[0009] In a first aspect, an embodiment of the present application provides a DPU-based peer-to-peer optical interconnection circuit switching networking method, the method comprising:

[0010] Collecting networking information of the optical interconnection circuit switching network, and establishing a peer-to-peer networking model according to the networking information, wherein the peer-to-peer networking model includes: a main network and an auxiliary network;

[0011] According to the peer-to-peer networking model, access nodes equipped with DPU accelerator cards are set up between the primary network and the secondary network for interconnection;

[0012] DPU adopts peer-to-peer traffic agreement to realize a peer-to-peer optical interconnection circuit interactive network.

[0013] In a specific embodiment of the present invention, according to the peer-to-peer networking model, an access node equipped with a DPU accelerator card is arranged between the primary network and the secondary network for interconnection and intercommunication, further comprising:

[0014] The DPU is connected to the auxiliary network through the first optical module cable to form a star topology data message exchange network;

[0015] The DPU is connected to the main network through the second optical module, and the optical fiber on the second optical module is directly connected to the main network port;

[0016] Any two DPUs are directly connected through the second optical module cable.

[0017] In a specific embodiment of the present invention, the above-mentioned DPU adopts a peer-to-peer traffic agreement to realize a peer-to-peer optical interconnection circuit interaction network, further comprising:

[0018] The same routing decision and main network configuration calculation algorithm is used inside any DPU;

[0019] Each time a DPU executes a routing reconstruction decision, the decision result is diffused by the auxiliary network to synchronize the routing status on all DPUs.

[0020] In a specific embodiment of the present invention, the above-mentioned DPU adopts a peer-to-peer traffic agreement including:

[0021] The DPU counts the inbound and outbound traffic demands from its own nodes, builds a priority queue, and decides on the current traffic priority processing without hindering the communication of other global nodes.

[0022] In a specific embodiment of the present invention, the above-mentioned DPU adopts a peer-to-peer traffic agreement including:

[0023] The time slice polling agreement divides a period of time into time slices corresponding to the number of DPUs, and stipulates that each DPU node has the highest routing decision-making power within its own time slice.

[0024] In a specific embodiment of the present invention, the DPU adopts a peer-to-peer traffic agreement and further includes:

[0025] The DPU sets up a credit lookup table, which records the credit value corresponding to each routing endpoint. After a route change occurs, the credit value corresponding to the changed routing endpoint is deducted. The credit value is restored after a period of time. The absolute value of the credit value is used to determine whether the current DPU has the right to directly conduct local routing arbitration without having to synchronize with other nodes before arbitration.

[0026] In a specific embodiment of the present invention, the above-mentioned DPU stores global traffic information, and traffic priority screening is performed locally in the DPU.

[0027] In a second aspect, an embodiment of the present application provides a DPU-based peer-to-peer optical interconnection circuit switching networking system, which adopts the above-mentioned DPU-based peer-to-peer optical interconnection circuit switching networking method, and the system includes:

[0028] Information collection module: used to collect networking information of the optical interconnection circuit switching network, and establish a peer-to-peer networking model based on the networking information, wherein the peer-to-peer networking model includes: a main network and an auxiliary network;

[0029] Based on DPU access node module: used to set up access nodes equipped with DPU accelerator cards between the primary network and the secondary network for interconnection according to the peer-to-peer networking model;

[0030] Peer-to-peer network building module: used for DPU to adopt peer-to-peer traffic agreement and realize peer-to-peer optical interconnection circuit interaction network.

[0031] In a third aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-mentioned DPU-based peer optical interconnection circuit switching networking method.

[0032] In a fourth aspect, an embodiment of the present application provides a DPU, comprising a memory, a dedicated acceleration unit, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the above-mentioned DPU-based peer optical interconnection circuit switching networking method are implemented.

[0033] Compared with the related prior art, it has the following outstanding beneficial effects:

[0034] 1) The method of the present invention proposes to use DPU on the access node side, so that the network no longer relies on the central switch arbitration unit, thereby avoiding the problem caused by the separation of the control plane and the data plane in the optical interconnection network;

[0035] 2) The method of the present invention proposes a peer-to-peer network so that all access nodes have the perception of the current state of the data plane and consistent routing derivation, a large number of routing judgments can be run locally, and the communication pressure of the control plane can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0037] Figure 1 The present invention is a schematic diagram of a peer-to-peer optical interconnection circuit switching network based on DPU;

[0038] Figure 2 This is a schematic diagram of a network configuration for a specific embodiment of the present invention;

[0039] Figure 3 This is a schematic diagram of a DPU accelerator card according to a specific embodiment of the present invention;

[0040] Figure 4 This is a networking diagram of the DPU accessing a hybrid network of the present invention;

[0041] Figure 5 This is a communication model diagram of a traditional composite network;

[0042] Figure 6 This is a flow chart of the route change request of the present invention;

[0043] Figure 7 The present invention is a schematic diagram of a peer-to-peer optical interconnection circuit switching networking system based on DPU. DETAILED DESCRIPTION

[0044] It should be noted that the processor described in the present invention is the control center of the electronic device, which may be a processor or a general term for multiple processing elements. For example, it may be one or more central processing units (CPUs), or application specific integrated circuits (ASICs), or one or more integrated circuits configured to implement the embodiments of the present invention, such as one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs).

[0045] Optionally, the processor can perform various functions of the electronic device by running or executing a software program stored in the memory, and calling data stored in the memory.

[0046] In a specific implementation, as an embodiment, the processor may include one or more CPUs. Each of these processors may be a single core processor or a multi-core processor. The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions). Electronic devices may include: servers, desktop computers, laptops, smart phones, tablet computers, embedded computers, etc., wherein the embedded computers include vehicles and robots, etc.

[0047] The memory is used to store the software program for executing the solution of the present invention, and the execution is controlled by the processor. The specific implementation method can refer to the above method embodiment and will not be repeated here.

[0048] It should be noted that the structure of the electronic device shown in the drawings of the present invention does not constitute a limitation thereto, and the actual knowledge structure recognition device may include more or fewer components than shown in the drawings, or a combination of certain components, or a different arrangement of components.

[0049] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0050] It should also be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0051] In the present invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0052] It should also be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0053] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0054] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0055] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0056] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0057] In order to make the above features and effects of the present invention more clearly and understandably described, embodiments are given below and described in detail with reference to the accompanying drawings. This specification discloses one or more embodiments that include the features of the present invention. The disclosed embodiments are only for illustration. The scope of protection of the present invention is not limited to the disclosed embodiments, and the present invention is defined by the attached claims.

[0058] The following is a system embodiment corresponding to the above method embodiment. This embodiment can be implemented in conjunction with the above embodiment. The relevant technical details mentioned in the above embodiment are still valid in this embodiment. In order to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this embodiment can also be applied in the above embodiment.

[0059] The present invention aims to propose a DPU-based peer-to-peer optical interconnection circuit switching networking method and system thereof, so as to reduce meaningless traffic of auxiliary networks, improve the communication efficiency of auxiliary networks, shorten the synchronization time window, and thus improve the global routing efficiency.

[0060] When conducting research on reconfigurable optical interconnection networks, the present invention found that the defects in the prior art were mainly caused by inappropriate access node side processing units. Existing access node side processing units tend not to handle complex routing problems but hand them over to the switch side for processing. However, as the scale of communication increases, handing all processing over to the switch side will bring serious communication overhead problems. After research, the present invention found that using DPU hardware on the access node side to handle routing problems and form a peer-to-peer network can effectively reduce the communication overhead of the auxiliary network, reduce the synchronization cost of routing information, shorten the delay, and thus improve the switching performance of the reconfigurable optical interconnection network.

[0061] The following is a detailed description with reference to specific embodiments:

[0062] Embodiment 1

[0063] like Figure 1 As shown, the embodiment of the present application provides a peer-to-peer optical interconnection circuit switching networking method based on DPU. The method of the present invention can be applied to the following methods:

[0064] Step 101, collecting networking information of the optical interconnection circuit switching network, and establishing a peer-to-peer networking model according to the networking information, wherein the peer-to-peer networking model includes: a primary network and an auxiliary network;

[0065] Step 102: according to the peer networking model, an access node equipped with a DPU accelerator card is set between the primary network and the secondary network for interconnection;

[0066] Step 103: The DPU adopts a peer-to-peer traffic agreement to implement a peer-to-peer optical interconnection circuit interactive network.

[0067] In a specific embodiment of the present invention, the above step 102, according to the peer networking model, sets an access node equipped with a DPU accelerator card between the primary network and the secondary network for interconnection, further comprising:

[0068] The DPU is connected to the auxiliary network through the first optical module cable to form a star topology data message exchange network; the DPU is connected to the main network through the second optical module, and the optical fiber on the second optical module is directly connected to the main network port; any two DPUs are directly connected through the second optical module cable.

[0069] like Figure 2 As shown, in a specific embodiment of the present invention, a peer-to-peer optical interconnection circuit switching network based on DPU, a feasible implementation example is Figure 2 As shown. The switching network is divided into two networks, the main network and the auxiliary network, which are not connected to each other. The main network is a reconfigurable optical interconnection network, such as a Crossbar network, a Benes network, etc., and has a driving circuit; the auxiliary network can be Ethernet. Usually, there is a huge bandwidth difference between the main network and the auxiliary network, and the bandwidth of the main network is more than 10 times that of the auxiliary network. The auxiliary network is only used to coordinate the routing decisions and changes of the configuration of the main network, and is not used to transmit data traffic.

[0070] In a specific embodiment of the present invention, the driver circuit board corresponding to the main network configuration is used to adjust the internal topology of the network. The driver circuit board may integrate a dedicated computing unit for calculating the mapping of a given routing request to the internal topology of the network, or may not integrate the above computing unit, but integrate it on the access node side. However, if integrated, the same computing unit needs to be implemented in the DPU on the access node side to ensure consistent behavior.

[0071] like Figure 3 As shown, in a specific embodiment of the present invention, the access node may use an accelerator card equipped with a DPU, which should provide two input and output interfaces, corresponding to the interface requirements of the primary network and the secondary network respectively. In a specific embodiment of the present invention, for a 400G optical interconnection network, its primary network interface may use a 400G QSFP56-DD optical module, and the secondary network interface may use a 10G SFP+ module, but the present invention is not limited thereto, and other types of optical modules may also be used.

[0072] Specifically, if Figure 4 As shown, in a specific embodiment of the present invention, the DPU is connected to the auxiliary network through a 10G SFP+ bidirectional optical module cable to form a traditional star topology data packet exchange network. The access method of the main network is slightly different. A 400G QSFP56-DD optical module is used on the DPU side, but when accessing the main network side, the optical fiber on the optical module is directly connected to the main network port. In terms of working principle, for any two connected DPU nodes, the main network is equivalent to a section of optical fiber, that is, the two DPU nodes can be regarded as being directly connected through a bidirectional 400G QSFP56-DD optical module cable without being forwarded through a switch.

[0073] In a specific embodiment of the present invention, in step 103, the DPU adopts a peer-to-peer traffic agreement to implement a peer-to-peer optical interconnection circuit interaction network, further comprising:

[0074] The same routing decision and main network configuration calculation algorithm is used inside any DPU;

[0075] Each time a DPU executes a routing reconstruction decision, the decision result is diffused by the auxiliary network to synchronize the routing status on all DPUs.

[0076] like Figure 5 As shown in the figure, the communication model of the traditional composite network. First, the access node side needs to initiate a routing request to the arbitration unit through the auxiliary network and wait for the result of the arbitration unit. Depending on the result, if the request is allowed, data communication will be directly initiated through the main network; otherwise, it will wait for a period of time and then initiate a routing request to the arbitration unit again. It is not difficult to find that for each request, the access node does not know or pay attention to the current global routing situation, nor does it know the decision logic of the arbitration unit, so it does not know which routes it requests will have a higher success rate, resulting in a large number of unallowed routing requests, that is, meaningless query traffic.

[0077] Among them, the arbitration unit is usually implemented by a dedicated ASIC chip or FPGA chip, and can also be implemented by a CPU or DPU. It is different from the routing query in the data packet exchange field. The arbitration unit technology is to solve routing conflicts. For example, if a→c and b→c are requested at the same time, it is necessary to decide (according to certain rules) which one is allowed and which one is rejected, because the two requests cannot be satisfied at the same time; the way to satisfy routing requests in this field is to establish corresponding point-to-point links, so for the same output port, there can only be one link at a time, and two cannot be established at the same time.

[0078] like Figure 6 As shown, the communication model of the technical solution of the present invention. The arbitration unit is no longer a single unit, but is distributed in each DPU node participating in the networking. Each DPU card uses the same routing decision and main network configuration calculation algorithm to ensure that the same result is derived when receiving the same input, achieving consistency that is not based on communication, thereby avoiding meaningless communication traffic caused by routing queries.

[0079] In a specific embodiment of the present invention, the above-mentioned DPU adopts a peer-to-peer traffic agreement including:

[0080] The DPU counts the inbound and outbound traffic demands from its own nodes, builds a priority queue, and decides on the current traffic priority processing without hindering the communication of other global nodes.

[0081] In a specific embodiment of the present invention, each time a node executes a routing reconstruction decision, the decision result is diffused by means of an auxiliary network to achieve synchronization of routing states.

[0082] In order to achieve more efficient global routing and avoid conflicts, the access node of the peer-to-peer network will adopt some conventions for network traffic characteristics. In the specific embodiment of the present invention, these conventions will be implemented in the DPU, using the DPU to count the inbound and outbound traffic requirements from its own node, build a priority queue and decide the traffic that should be processed first and least hinder the communication of other global nodes. Compared with the traditional centralized solution, the DPU on the access node side in this solution has global traffic information, so the above-mentioned traffic priority screening work can be performed locally instead of on the switch side, avoiding additional communication traffic with the traditional switch and the meaningless communication traffic that may be generated.

[0083] In a specific embodiment of the present invention, the above-mentioned DPU adopts a peer-to-peer traffic agreement including:

[0084] The time slice polling agreement divides a period of time into time slices corresponding to the number of DPUs, and stipulates that each DPU node has the highest routing decision-making power within its own time slice.

[0085] In a specific embodiment of the present invention, a typical agreement is a time slice polling agreement, which divides a period of time (usually several to dozens of μs) into N time slices, where N is the number of networked DPUs. It is agreed that each DPU node has the highest decision-making power within its own time slice, thereby avoiding additional query-waiting result process within the time slice and also avoiding the corresponding meaningless communication traffic.

[0086] In a specific embodiment of the present invention, the DPU adopts a peer-to-peer traffic agreement and further includes:

[0087] The DPU sets up a credit lookup table, which records the credit value corresponding to each routing endpoint. After a route change occurs, the credit value corresponding to the changed routing endpoint is deducted. The credit value is restored after a period of time. The absolute value of the credit value is used to determine whether the current DPU has the right to directly conduct local routing arbitration without having to synchronize with other nodes before arbitration.

[0088] Furthermore, a credit lookup table can be deployed on the DPU side. A simple implementation is: the node with the highest credit value has the priority to choose the current route endpoint; if the node gives up the choice, it will be filled down according to the credit value. In this way, route arbitration can be completed within the DPU. After the route is changed, the credit value corresponding to the route endpoint after the change is deducted according to the expected communication time, and then gradually restored after a period of time. According to the absolute value of the credit value, it is determined whether it is currently entitled to directly conduct local arbitration without having to synchronize with other nodes before arbitration, which can avoid some meaningless communication traffic.

[0089] The credit lookup table can be described by a matrix, where each row corresponds to one of the starting nodes and each column corresponds to one of the routing endpoints. The value obtained by selecting the row and column is the credit value from the corresponding starting point to the endpoint. The node's own credit value increases regularly through a specific algorithm, and the credit values ​​of the remaining rows are synchronized through the communication of the auxiliary network.

[0090] In a specific embodiment of the present invention, the above-mentioned DPU stores global traffic information, and traffic priority screening is performed locally in the DPU.

[0091] When the switch side chooses to integrate a dedicated hardware unit with the same behavior as the access node side, the communication traffic of the auxiliary network can be further reduced: the information transmission to the driving circuit is simplified from specific network configuration information to route change information, and the corresponding network configuration information is calculated by the dedicated hardware unit on the switch side and passed to the driving circuit to complete the main network configuration.

[0092] Embodiment 2

[0093] like Figure 7 As shown, the embodiment of the present application provides a DPU-based peer-to-peer optical interconnection circuit switching networking system, which adopts the above-mentioned DPU-based peer-to-peer optical interconnection circuit switching networking method, and the system includes:

[0094] The information collection module 201 is used to collect the networking information of the optical interconnection circuit switching network, and establish a peer-to-peer networking model according to the networking information, wherein the peer-to-peer networking model includes: a primary network and an auxiliary network;

[0095] DPU-based access node module 202: used to set access nodes equipped with DPU acceleration cards between the primary network and the secondary network for interconnection according to the peer-to-peer networking model;

[0096] Peer-to-peer network building module 203: used for the DPU to adopt peer-to-peer traffic agreement to realize a peer-to-peer optical interconnection circuit interactive network.

[0097] Embodiment 3

[0098] An embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of the above-mentioned DPU-based peer-to-peer optical interconnection circuit switching networking method are implemented.

[0099] Embodiment 4

[0100] like Figure 3 As shown, an embodiment of the present application provides a DPU, including a memory, a dedicated acceleration unit, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the above-mentioned DPU-based peer optical interconnection circuit switching networking method are implemented.

[0101] In summary, compared with the prior art, the present invention is as follows: the existing circuit switching network usually uses a single centralized control node to complete global routing decisions and network status control. When the number of network access nodes is large, the auxiliary network will bear a large amount of meaningless communication traffic caused by node query routing, which will reduce the performance of the auxiliary network and affect the reconstruction efficiency of the main network. After using this solution, each access node maintains a global network routing model locally, and moves the routing query to the local area, thereby avoiding meaningless communication traffic, reducing the pressure on the auxiliary network, and increasing the available reconstruction frequency of the network, thereby improving the network switching performance.

[0102] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0103] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A peer-to-peer optical interconnection circuit switching networking method based on DPU, characterized in that: The method comprises: Collecting networking information of an optical interconnection circuit switching network, and establishing a peer-to-peer networking model according to the networking information, wherein the peer-to-peer networking model includes: a primary network and an auxiliary network; According to the peer-to-peer networking model, an access node equipped with a DPU accelerator card is arranged between the primary network and the secondary network for interconnection; The DPU adopts a peer-to-peer traffic agreement to realize a peer-to-peer optical interconnection circuit interaction network.

2. The DPU-based peer-to-peer optical interconnection circuit switching networking method according to claim 1, characterized in that: According to the peer-to-peer networking model, between the primary network and the secondary network, an access node equipped with a DPU accelerator card is arranged for interconnection and intercommunication, further comprising: The DPU is connected to the auxiliary network via a first optical module cable to form a star topology data message exchange network; The DPU is connected to the main network through a second optical module, and the optical fiber on the second optical module is directly connected to the main network port; Any two of the DPUs are directly connected via the second optical module cable.

3. The DPU-based peer-to-peer optical interconnection circuit switching networking method according to claim 1, characterized in that: The DPU adopts a peer-to-peer traffic agreement to implement a peer-to-peer optical interconnection circuit interaction network, further comprising: Any of the DPUs uses the same routing decision and main network configuration calculation algorithm; Each time when the DPU executes a routing reconstruction decision, the auxiliary network diffuses the decision result to achieve synchronization of routing states on all the DPUs.

4. The DPU-based peer-to-peer optical interconnection circuit switching networking method according to claim 1, characterized in that: The DPU adopts a peer-to-peer traffic agreement including: The DPU counts the inbound and outbound traffic demands from its own nodes, builds a priority queue and decides on the current traffic priority processing without hindering the traffic of other global nodes' communication.

5. The DPU-based peer-to-peer optical interconnection circuit switching networking method according to claim 1, characterized in that: The DPU adopts a peer-to-peer traffic agreement including: The time slice polling agreement divides a period of time into time slices corresponding to the number of DPUs, and stipulates that each DPU node has the highest routing decision-making power within its own time slice.

6. The DPU-based peer-to-peer optical interconnection circuit switching networking method according to claim 5, characterized in that: The DPU adopts a peer-to-peer traffic agreement that also includes: The DPU sets a credit lookup table, which records the credit value corresponding to each routing endpoint. After a routing change occurs, the credit value corresponding to the changed routing endpoint is deducted, and the credit value is restored after a period of time. According to the absolute value of the credit value, it is determined whether the current DPU has the right to directly conduct local routing arbitration without having to synchronize with other nodes before arbitration.

7. The DPU-based peer-to-peer optical interconnection circuit switching networking method according to claim 4, characterized in that: The DPU stores global traffic information, and traffic priority screening is performed locally on the DPU.

8. A DPU-based peer-to-peer optical interconnection circuit switching networking system, using the DPU-based peer-to-peer optical interconnection circuit switching networking method as claimed in any one of claims 1 to 7, characterized in that: The system comprises: An information collection module is used to collect networking information of an optical interconnection circuit switching network, and establish a peer-to-peer networking model according to the networking information, wherein the peer-to-peer networking model includes: a primary network and an auxiliary network; DPU access node module: used to set an access node equipped with a DPU accelerator card between the primary network and the auxiliary network for interconnection according to the peer networking model; Peer-to-peer network building module: used for the DPU to adopt peer-to-peer traffic agreement to realize the peer-to-peer optical interconnection circuit interaction network.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the DPU-based peer-to-peer optical interconnection circuit switching networking method described in any one of claims 1 to 7 are implemented.

10. A DPU comprising a memory, a dedicated acceleration unit, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the DPU-based peer-to-peer optical interconnection circuit switching networking method are implemented as described in any one of claims 1 to 7.

Images