A method and device for software-configurable data packet switching

Abstract

The present invention provides a method and device for exchanging data packets based on software configurability, including: a sending module sends a data packet accompanied by a virtual port signal; a command query request is initiated to a command configuration module according to the virtual port signal; the command configuration module queries the corresponding command in its command query table according to the virtual port, and sends the command to a command issuing module; the command issuing module obtains the ID number assigned to the data packet from an ID allocation and recovery module, and sends the command and the ID number of the data packet to each receiving module; each receiving module determines whether the command belongs to itself, and if so, it is used as a destination module to retain the command, otherwise it is directly discarded; a network chip interconnection module establishes a data path between a sending module and a destination module based on the ID number; after the destination module receives the data packet, the ID number is released, and the ID number is sent to the ID allocation and recovery module. The method provided by the present invention can flexibly realize data packet exchange, support multi-functional DPU scenarios, and has strong scalability.

Description

Technical Field

[0001] The present invention relates to the technical field of data packet interaction, and in particular to a software-configurable data packet exchange method and device. Background Art

[0002] The Data Processing Unit (DPU) is a new generation of processor for data-centric computing. It integrates complete data center functions on a single chip. Together with the Central Processing Unit (CPU) and the Graphics Processing Unit (GPU), it constitutes the three pillars of new computing.

[0003] At present, DPU chips can support a variety of business scenarios. The design modules of various business scenarios need to interact with a large number of data packets. The traditional design idea is to add metadata to multiple data streams, such as a field or domain, to mark the destination module for recording data, and then distribute the data stream to the corresponding module for processing according to the value of the added field or domain after data aggregation. This method is relatively simple to implement in a single business scenario mode, but it is inadequate for a DPU scenario that supports multiple functions.

[0004] Specifically, the following problems exist:

[0005] Occupied bandwidth. This processing method will affect the bandwidth of small packets (i.e. smaller data packets in the data stream). Generally, the length of a small packet is 64 bytes. If the data path is designed according to the data bit width of 256 bits, 2 beats of data are required, that is, 2 clock cycles are required to transmit the complete small packet data; if 1 beat of data is added by adding metadata, the clock cycle required for data transmission will increase to 3 beats, which will waste 1 / 3 of the data bandwidth.

[0006] The scalability is poor. If the business changes, the data flow destination module needs to be modified, and this modification can only be achieved by modifying the code. Its scalability cannot meet the requirements of rapid implementation. Summary of the invention

[0007] In view of this, an embodiment of the present invention provides a software-configurable data packet switching method and device to eliminate or improve one or more defects existing in the prior art and solve the problems existing in the existing data packet interaction solutions, such as not supporting multiple business scenarios, occupying and wasting bandwidth, and poor scalability.

[0008] On the one hand, the present invention provides a software-configurable data packet switching method, the method is executed in a data processing unit, the data processing unit includes an instruction configuration module, an ID allocation and recovery module, an instruction issuing module and a network chip interconnection module, and the method includes the following steps:

[0009] The sending module sends a data packet and sends a virtual port signal at the same time; and initiates an instruction query request to the instruction configuration module according to the virtual port signal;

[0010] The instruction configuration module queries the instruction of the data packet in its instruction query table according to the corresponding virtual port, and sends the instruction of the data packet to the instruction issuing module;

[0011] After receiving the instruction, the instruction issuing module obtains the ID number assigned to the data packet from the ID allocation and recovery module; the instruction issuing module sends the instruction and the ID number of the data packet to each receiving module;

[0012] Each receiving module determines whether the instruction belongs to itself. If so, the receiving module is used as the destination module and the instruction is retained. If not, the instruction is directly discarded.

[0013] The network chip interconnection module establishes a data path between the sending module and the destination module based on the ID number, and the sending module sends the data packet to the destination module;

[0014] After receiving the data packet, the destination module releases the ID number and sends the ID number to the ID allocation and recovery module.

[0015] In some embodiments of the present invention, initiating an instruction query request to the instruction configuration module according to the virtual port signal further includes:

[0016] After receiving the instruction query request, the instruction configuration module responds in a multi-way polling arbitration manner.

[0017] In some embodiments of the present invention, the instruction configuration module queries the instruction of the data packet in its instruction query table according to the corresponding virtual port, and further includes:

[0018] The instruction configuration module sends the virtual port as an address to the instruction query table, and the instruction query table queries the corresponding instruction according to the address.

[0019] In some embodiments of the present invention, the method further comprises:

[0020] The instruction issuing module transmits the instruction and ID number of the data packet to each receiving module in a broadcasting manner.

[0021] In some embodiments of the present invention, after receiving the instruction, the instruction issuing module obtains the ID number allocated to the data packet from the ID allocation and recovery module, and further includes:

[0022] In the ID allocation and recovery unit, if the ID number resources have been allocated, it is necessary to wait until the ID number resources are released.

[0023] In some embodiments of the present invention, each receiving module determines whether the instruction belongs to itself, further comprising:

[0024] The receiving module receives and parses the instructions of the data packet, obtains the IP index field in the instruction, and determines whether the IP index field is consistent with the IP index defined by its own module. If they are consistent, it belongs to itself and serves as the destination module and retains the instruction. If they are inconsistent, it does not belong to itself and directly discards the instruction.

[0025] In some embodiments of the present invention, the network chip interconnection module establishes a data path between the sending module and the destination module based on the ID number, and further includes:

[0026] The network chip interconnection module uses the same ID number of the sending module and the destination module as an identifier and establishes a data path according to a router.

[0027] On the other hand, the present invention provides a software-configurable data packet switching device, the device is applied to a data processing unit, and the device includes:

[0028] An instruction configuration module is used to configure the data packets of each module based on software and generate an instruction query table; according to the virtual port signal when the sending module sends the data packet, the instruction of the corresponding data packet is obtained from the instruction query table;

[0029] An ID allocation and recovery module is used to allocate an ID number to each data packet and recover the ID number of the data packet that has completed data transmission;

[0030] An instruction issuing module, used for sending the instruction of the corresponding data packet obtained from the instruction configuration module and the ID number of the corresponding data packet obtained from the ID allocation and recovery module to each receiving module to determine the destination module;

[0031] A network chip interconnection module is used to establish a data path between the sending module and the destination module based on the ID number.

[0032] In some embodiments of the present invention, the device further comprises an alarm module for providing an alarm prompt for abnormalities occurring in the instruction configuration module, the ID allocation and recovery module, the instruction issuing module and the network chip interconnection module.

[0033] On the other hand, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the methods mentioned above.

[0034] The beneficial effects of the present invention are at least:

[0035] The present invention provides a method and device for exchanging data packets based on software configurability, including: a sending module sends a data packet accompanied by a virtual port signal; an instruction query request is initiated to an instruction configuration module according to the virtual port signal; the instruction configuration module queries the corresponding instruction in its instruction query table according to the virtual port, and sends the instruction to the instruction issuing module; the instruction issuing module obtains the ID number assigned to the data packet from the ID allocation and recovery module, and sends the instruction and the ID number of the data packet to each receiving module together; each receiving module determines whether the instruction belongs to itself, and if so, it is used as a destination module to retain the instruction, otherwise it is directly discarded; the network chip interconnection module establishes a data path between the sending module and the destination module based on the ID number; after the destination module receives the data packet, the ID number is released, and the ID number is sent to the ID allocation and recovery module. The method provided by the present invention is based on software configuration data packet information, can flexibly realize data packet exchange, support multi-functional DPU scenarios, and has strong scalability.

[0036] Additional advantages, purposes, and features of the present invention will be described in part in the following description, and will become apparent to those skilled in the art after studying the following, or may be learned from the practice of the present invention. The purposes and other advantages of the present invention may be achieved and obtained by the structures specifically indicated in the specification and the accompanying drawings.

[0037] Those skilled in the art will appreciate that the objectives and advantages that can be achieved with the present invention are not limited to the above specific description, and the above and other objectives that can be achieved by the present invention will be more clearly understood from the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of the present application, and do not constitute a limitation of the present invention. In the drawings:

[0039] Figure 1 The figure is a schematic diagram of the steps of a software-configurable data packet switching method in one embodiment of the present invention.

[0040] Figure 2The figure is a schematic diagram of the structural flow of a software-configurable data packet switching method in one embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0042] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the solutions according to the present invention are shown in the accompanying drawings, while other details that are not closely related to the present invention are omitted.

[0043] It should be emphasized that the term “include / comprises” when used herein refers to the presence of features, elements, steps or components, but does not exclude the presence or addition of one or more other features, elements, steps or components.

[0044] It should also be noted that, unless otherwise specified, the term “connection” herein may refer not only to a direct connection but also to an indirect connection involving an intermediate.

[0045] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent the same or similar components, or the same or similar steps.

[0046] It should be emphasized here that the step marks mentioned below are not intended to limit the order of the steps, but it should be understood that the steps can be executed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be executed simultaneously.

[0047] In order to solve the problems of existing data packet interaction solutions that do not support multiple business scenarios, occupy and waste bandwidth, and have poor scalability, the present invention proposes a software-configurable data packet exchange method, which is executed in a data processing unit DPU, wherein the data processing unit includes an instruction configuration module, an ID allocation and recovery module, an instruction issuing module, and a network chip (Network on Chip, NOC) interconnection module, such as Figure 1 As shown, the method includes the following steps S101 to S106:

[0048] Step S101: The sending module sends a data packet and sends a virtual port signal at the same time; and initiates an instruction query request to the instruction configuration module according to the virtual port signal.

[0049] Step S102: the instruction configuration module queries the instruction of the data packet in its instruction query table according to the corresponding virtual port, and sends the instruction of the data packet to the instruction issuing module.

[0050] Step S103: After receiving the instruction, the instruction issuing module obtains the ID number assigned to the data packet from the ID allocation and recovery module; the instruction issuing module sends the instruction and the ID number of the data packet to each receiving module.

[0051] Step S104: Each receiving module determines whether the instruction belongs to itself. If so, the receiving module is used as the destination module and the instruction is retained. If not, the instruction is directly discarded.

[0052] Step S105: the network chip interconnection module establishes a data path between the sending module and the destination module based on the ID number, and the sending module sends the data packet to the destination module.

[0053] Step S106: After receiving the data packet, the destination module releases the ID number and sends the ID number to the ID allocation and recovery module.

[0054] like Figure 2 The figure shows a structural flow chart of a software-configurable data packet switching method.

[0055] In step S101, the sending module sends a data packet accompanied by a virtual port (vport) signal, and initiates an instruction query request to the instruction configuration module based on the virtual port signal.

[0056] For example, Figure 2 As shown in the process (1), the sending module is module 1, and module 1 initiates an instruction query request to the instruction configuration module based on the virtual port signal.

[0057] In step S102, the instruction configuration module receives the instruction query request, uses the virtual port as the address, searches for the corresponding instruction in the instruction query table based on the virtual port address, and sends the searched instruction to the instruction issuing module. Step S102 corresponds to Figure 2 (2) Process.

[0058] In some embodiments, there may be multiple sending modules that need to send data packets, thereby issuing instruction query requests. Therefore, when the instruction configuration module receives the instruction query request, the instruction query request is first stored in the request queue and responded to by multi-way polling arbitration. The arbitrator will periodically poll the request queue and select the next request to respond according to a certain priority strategy. Exemplarily, different strategies such as fixed priority, rotating priority or weight-based priority can be used for arbitration. The use of multi-way polling arbitration can coordinate the access order between multiple requests, thereby avoiding conflicts and competition between different requests. By responding to requests from different modules in turn, fair access and resource sharing can be achieved, thereby improving the overall efficiency and performance of the system.

[0059] In some embodiments, in the instruction configuration module, information such as instructions and data flow of the data packet is pre-set by software.

[0060] In step S103, after receiving the instruction corresponding to the data packet, the instruction issuing module obtains the unique ID number assigned to the data packet from the ID allocation and recovery module, and sends the instruction and ID number of the data packet to the receiving module. Step S103 corresponds to Figure 2 Processes (3) and (4).

[0061] In some embodiments, during the period of obtaining the unique ID number assigned to the data packet from the ID allocation and recovery module, if the ID number resources in the ID allocation and recovery module have been allocated, it is necessary to wait until an ID number is released and allocate the released ID number to the data packet. At the same time, it should be noted that when the ID allocation and recovery module has no ID number resource allocation, the instruction issuing module will not send the instruction of the data packet to the receiving module alone, and only send the instruction of the data packet and the ID number to the receiving module when the ID number of the data packet is obtained.

[0062] In some embodiments, the instruction issuing module transmits the instruction and ID number of the data packet to each receiving module in a broadcasting manner. Broadcasting can send the instruction and ID number of the data packet to the modules in the entire network, rather than to a specific module.

[0063] In step S104, each receiving module receives the instruction and ID number of the data packet, parses the instruction, and determines whether the instruction belongs to its own module. If it does, it is used as the destination module and the instruction is retained. If it does not, the instruction is directly discarded.

[0064] In some embodiments, each receiving module parses the instruction, obtains the IP index field in the instruction, and compares the obtained IP index field with the IP index defined by its own module, wherein each module has a unique IP index, which is pre-set before execution. If the obtained IP index field is consistent with the IP index defined by its own module, it means that the instruction belongs to this module, and this module serves as the destination module of the data packet and retains the instruction; if the obtained IP index field is inconsistent with the IP index defined by its own module, it means that the instruction does not belong to this module and the instruction can be directly discarded. For example, Figure 2 As shown in , after the determination by each receiving module, module 5 is the destination module. At this time, module 5 retains the instruction of the data packet and the ID number of the data packet received through broadcasting.

[0065] In step S105, the sending module and the destination module both have the ID number of the data packet. The network chip interconnection module establishes a data path between the sending module and the destination module based on the ID number and the path determined by the router or parser to ensure that the data packet can be transmitted from the sending module to the destination module.

[0066] Exemplarily, step S105 corresponds to Figure 2 In process (5), the network chip interconnection module establishes a data path between module 1 and module 5 based on the ID number.

[0067] Among them, the network chip NOC is a network structure implemented at the chip level, which adopts a certain topology and communication protocol to connect various functional modules and processing units inside the chip and provide high-speed, low-latency data transmission channels.

[0068] In step S106, based on the data path established in step S105, the sending module sends the data packet to the destination module. After the destination module completely receives the data packet, it releases the ID number of the data packet and sends the ID number to the ID allocation and recovery module for recovery so that the ID allocation and recovery module can allocate and use the ID number again. Step S106 corresponds to Figure 2 (6) Process.

[0069] Corresponding to the software-configurable data packet switching method, the present invention also provides a software-configurable data packet switching device, which is applied to a data processing unit and includes the following modules:

[0070] The instruction configuration module is used to configure the data packets of each module based on software and generate an instruction query table; according to the virtual port signal when the sending module sends the data packet, the instruction of the corresponding data packet is obtained from the instruction query table.

[0071] The ID allocation and recycling module is used to allocate an ID number to each data packet and recycle the ID number of the data packet that has completed data transmission.

[0072] The instruction issuing module is used to send the instruction of the corresponding data packet obtained from the instruction configuration module and the ID number of the corresponding data packet obtained from the ID allocation and recovery module to each receiving module to determine the destination module.

[0073] The network chip interconnection module is used to establish a data path between the sending module and the destination module based on the ID number to realize data transmission between the sending module and the destination module.

[0074] In some embodiments, the software-configurable data packet switching device also includes an alarm module for issuing an alarm prompt for abnormalities occurring in the instruction configuration module, the ID allocation and recovery module, the instruction issuing module, and the network chip interconnection module.

[0075] The present invention also 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 software-configurable data packet switching method are implemented.

[0076] Corresponding to the above method, the present invention also provides a device, which includes a computer device, the computer device includes a processor and a memory, the memory stores computer instructions, the processor is used to execute the computer instructions stored in the memory, and when the computer instructions are executed by the processor, the device implements the steps of the method described above.

[0077] The embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the aforementioned edge computing server deployment method are implemented. The computer-readable storage medium can be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field.

[0078] In summary, the present invention provides a method and device for data packet switching based on software configurability, including: a sending module sends a data packet accompanied by a virtual port signal; an instruction query request is initiated to an instruction configuration module according to the virtual port signal; the instruction configuration module queries the corresponding instruction in its instruction query table according to the virtual port, and sends the instruction to the instruction issuing module; the instruction issuing module obtains the ID number assigned to the data packet from the ID allocation and recovery module, and sends the instruction and the ID number of the data packet to each receiving module together; each receiving module determines whether the instruction belongs to itself, and if so, it is used as a destination module to retain the instruction, otherwise it is directly discarded; the network chip interconnection module establishes a data path between the sending module and the destination module based on the ID number; after the destination module receives the data packet, it releases the ID number and sends the ID number to the ID allocation and recovery module. The method provided by the present invention is based on software configuration data packet information, can flexibly realize data packet switching, supports multi-functional DPU scenarios, and has strong scalability.

[0079] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0080] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0081] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with features of other embodiments or replace features of other embodiments.

[0082] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the embodiments of the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A software-configurable data packet switching method, characterized in that: The method is executed in a data processing unit, which includes an instruction configuration module, an ID allocation and recovery module, an instruction issuing module and a network chip interconnection module. The method includes the following steps: The sending module sends a data packet and sends a virtual port signal at the same time; and initiates an instruction query request to the instruction configuration module according to the virtual port signal; The instruction configuration module queries the instruction of the data packet in its instruction query table according to the corresponding virtual port, and sends the instruction of the data packet to the instruction issuing module; After receiving the instruction, the instruction issuing module obtains the ID number assigned to the data packet from the ID allocation and recovery module; the instruction issuing module sends the instruction and the ID number of the data packet to each receiving module; Each receiving module receives and parses the instruction of the data packet, obtains the IP index field in the instruction, and determines whether the IP index field is consistent with the IP index defined by its own module. If they are consistent, it belongs to itself, and the receiving module is used as the destination module to retain the instruction. If they are inconsistent, it does not belong to itself and the instruction is directly discarded. The network chip interconnection module establishes a data path between the sending module and the destination module based on the ID number, and the sending module sends the data packet to the destination module; After receiving the data packet, the destination module releases the ID number and sends the ID number to the ID allocation and recovery module.

2. The software-configurable data packet switching method according to claim 1, characterized in that: Initiating an instruction query request to the instruction configuration module according to the virtual port signal also includes: After receiving the instruction query request, the instruction configuration module responds in a multi-way polling arbitration manner.

3. The software-configurable data packet switching method according to claim 1, characterized in that: The instruction configuration module queries the instruction of the data packet in its instruction query table according to the corresponding virtual port, and also includes: The instruction configuration module sends the virtual port as an address to the instruction query table, and the instruction query table queries the corresponding instruction according to the address.

4. The software-configurable data packet switching method according to claim 1, characterized in that: The method further comprises: The instruction issuing module transmits the instruction and ID number of the data packet to each receiving module in a broadcasting manner.

5. The software-configurable data packet switching method according to claim 1, characterized in that: After receiving the instruction, the instruction issuing module obtains the ID number allocated by the data packet from the ID allocation and recovery module, and further includes: In the ID allocation and recovery unit, if the ID number resources have been allocated, it is necessary to wait until the ID number resources are released.

6. The software-configurable data packet switching method according to claim 1, characterized in that: The network chip interconnection module establishes a data path between the sending module and the destination module based on the ID number, and further includes: The network chip interconnection module uses the same ID number of the sending module and the destination module as an identifier and establishes a data path according to a router.

7. A software-configurable data packet switching device, characterized in that: The device is applied to a data processing unit, and comprises: An instruction configuration module is used to configure the data packets of each module based on software and generate an instruction query table; according to the virtual port signal when the sending module sends the data packet, the instruction of the corresponding data packet is obtained from the instruction query table; An ID allocation and recovery module is used to allocate an ID number to each data packet and recover the ID number of the data packet that has completed data transmission; An instruction issuing module, used for sending the instruction of the corresponding data packet obtained from the instruction configuration module and the ID number of the corresponding data packet obtained from the ID allocation and recovery module to each receiving module to determine the destination module, including each receiving module receiving and parsing the instruction of the data packet, obtaining the IP index field in the instruction, judging whether the IP index field is consistent with the IP index defined by its own module, if consistent, it belongs to itself, and the receiving module is used as the destination module to retain the instruction, if inconsistent, it does not belong to itself, and the instruction is directly discarded; A network chip interconnection module is used to establish a data path between the sending module and the destination module based on the ID number.

8. The software-configurable data packet switching device according to claim 7, characterized in that: The device also includes an alarm module, which is used to issue an alarm prompt for abnormalities occurring in the instruction configuration module, the ID allocation and recovery module, the instruction issuing module and the network chip interconnection module.

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 method according to any one of claims 1 to 6 are implemented.

Images