Processor-reconfigurable programmable switching structure and programmable data plane chip
By designing a pipeline or RTC processor for reconfigurable processors in a programmable switching structure and processing data packets in parallel, the problems of high packet processing delay and insufficient throughput in the prior art are solved, and efficient network service performance is achieved and cost-reduced data center networks.
Patent Information
- Application Number
- CN202410929824.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-11
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-07-11
AI Technical Summary
When existing programmable switching chips process packets with a large number of computing requirements, long processing flow and dependencies, the pipeline architecture needs to recycle the packets, resulting in increased latency, reduced throughput and state consistency difficult to ensure. The existing solution The solution is expensive and cannot completely solve the problem.
Design a programmable switching structure for processor reconfigurable, and reconstruct the reconfigurable processor into a pipeline stage or RTC processor to realize flexible transmission of data packet header vectors and other data, allowing each reconfigurable processor to be Reconstructed into a pipeline or RTC processor, and connects through data paths to form a complete pipeline or an independent RTC processor, supporting parallel processing of multi-core RTC cores.
In the case of consuming a small amount of chip area, the programmability of the switch is greatly improved, network communication delays are reduced, network service performance is improved, the demand for servers and intermediate box equipment is reduced, and the construction and maintenance costs of data center networks are solved, and architectural programming is solved Problems of sexual limitations, insufficient throughput and high latency.
Smart Images

Figure CN119071255A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of programmable data planes, and in particular to a programmable switching structure and a programmable data plane chip with reconfigurable processors. Background Art
[0002] Software-defined networks include data planes and control planes. The data plane is responsible for protocol parsing and reverse parsing, and packet processing, while the control plane is responsible for issuing parsing rules and flow table matching rules. From customized protocol forwarding to supporting in-network computing applications, high-performance programmable switching chips have further explored the potential of data plane devices. The architecture of programmable switching chips can be divided into three types: pipeline, RTC (Run-To-Complete), and the fusion architecture of the two.
[0003] Programmable switching chips using pipeline architecture often have relatively high throughput and relatively low processing latency, but they have the following disadvantages:
[0004] 1. Each pipeline stage can perform a limited number of computational steps;
[0005] 2. Due to the unidirectional nature of the pipeline, there is no efficient mechanism to pass data to the early stages;
[0006] In response to the above shortcomings, if the packet processing process meets any of the following conditions: it contains a large number of computing requirements, the processing process is long and cannot be parallelized due to dependencies, or the state data needs to be written back to the early stage, then the pipeline architecture needs to recirculate the data packet. Recirculation will cause the latency of the recirculated data packet to increase significantly, the system throughput to decrease, and the recirculated data packet to occupy the pipeline processing capacity multiple times, thereby reducing the background traffic processing capacity other than the recirculated traffic and generating potential disorder, thereby destroying the strong state consistency of stateful applications.
[0007] If recirculation is not used, there are generally three solutions:
[0008] 1. Submit such tasks to the control plane for processing;
[0009] 2. Forward the data packets that need to perform such tasks to the dedicated network function server for processing;
[0010] 3. Connect an FPGA (Field-Programmable Gate Array) or other dedicated hardware processing to the switch chip;
[0011] The above three solutions all require a significant increase in the cost of the solutions, and still cannot solve the problem of significantly increased latency of the recirculated data packets and potential disorder, thereby destroying the strong state consistency of stateful applications.
[0012] The RTC processor can process a data packet until the entire processing flow is completed, and can execute longer and more complex processing flows without forwarding it to other hardware modules. However, it also faces problems such as insufficient throughput, high latency, and difficulty in ensuring state consistency. Summary of the invention
[0013] The present application provides a programmable switching structure with a reconfigurable processor and a programmable data plane chip to solve the problems of limited architectural programmability, insufficient throughput, and high latency in network communications of programmable switching chips in related technologies.
[0014] The first aspect of the present application provides a processor-reconfigurable programmable switching structure, including: a first side path for transmitting PHV (Packet Header Vector); a second side path for transmitting other data and control signals other than PHV; a reconfigurable processor sequence, the reconfigurable processor sequence includes multiple reconfigurable processors, the first side path and the second side path connect the multiple reconfigurable processors, wherein each reconfigurable processor of the multiple reconfigurable processors is allowed to be reconfigured into a pipeline stage or run to complete an RTC processor, and there is a data path for transmitting PHV between each adjacent reconfigurable processor, when both ends of the data path are reconfigurable processors reconfigured into pipeline stages, the data path is opened to connect the two pipeline stages to a complete pipeline, and when both ends of the data path are not reconfigurable processors reconfigured into pipeline stages, the data path is closed. The reconfigurable processor reconfigured into an RTC processor is connected via the first side path.
[0015] Optionally, the reconfigurable processor includes a processor in which all of the following items are reconfigurable, or a processor in which some of the following items are reconfigurable and the other items are not included in the processor, or a processor in which some of the following items are reconfigurable and the other items are not reconfigurable: a register array, in the pipeline stage, the register array is used to store data packet header fields, metadata and intermediate signals, and in the RTC processor, the register array serves as a register group specified by the instruction system; a data memory, in the pipeline stage, the data memory is used to store lookup tables and state tables, the lookup tables are used for lookups when data packets pass through, and the state tables are used for reading, modifying and writing back states when data packets pass through, in the RTC processor, the data memory is used to store data packet header fields, metadata, lookup tables and state tables, and intermediate signals, and if the RTC processor is equipped with an operating system, the data memory is also used to store data structures related to operating system maintenance; an instruction memory, used to store instructions in both the pipeline stage and the RTC processor; an ALU (Arithmetic and Logic Unit), used to perform arithmetic and logical calculations in both the pipeline stage and the RTC processor; a SALU (Stateful Arithmetic and Logic Unit) Unit, arithmetic logic unit with state), is used to read the state from the state table of the data storage in the pipeline stage, and write the calculation result back to the state table after performing arithmetic logic calculation; in the RTC processor, it is used to read data from the data storage, and write the calculation result back to the data storage after performing arithmetic logic calculation; the matching module is used in both the pipeline stage and the RTC processor to search and match in the lookup table of the data storage according to the given keyword, and return whether there is a matching table entry and the corresponding table entry content.
[0016] Optionally, when the reconfigurable processor is reconfigured into a pipeline stage, it includes: after the PHV of the data packet is received by the current processor, it is written into the register array of the current processor; according to the keyword selection rule specified by the user program, a part of the fields is selected to form the keyword and transmitted to the matching module; the matching module searches and matches in the lookup table of the data storage according to the given keyword, and when there is a matching table entry, the execution operation ID (IDentification) is determined according to the content of the table entry; when there is no matching table entry, the default operation ID specified by the user program is used; according to the operation ID, the instruction is read from the instruction memory, the instruction is parsed using a decoder, and the parsed instruction is dispatched to the ALU and SALU, wherein the ALU reads the fields involved in the calculation from the register array or the instruction parameter, and writes back to the register array after calculation; the SALU reads the fields involved in the calculation from the register array or the instruction parameter, and reads the status involved in the calculation from the status table in the data storage, and writes back to the register array and / or the status table in the data storage after calculation; in the processing flow, the PHV of a data packet is transmitted in the registers of the register array corresponding to different pipeline beats, and after the processing is completed, the corresponding PHV is sent out of the current processor.
[0017] Optionally, when the reconfigurable processor is reconstructed into an RTC processor, it includes: after the PHV of the data packet is received by the current processor, it is written into the data memory of the current processor, and the current processor is notified to start executing the processing flow, and the current processor reads the instruction to be executed from the instruction register according to the program counter until the processing termination instruction is read; each instruction read is parsed by a dedicated decoder and dispatched to the ALU or SALU, wherein the ALU reads the fields involved in the calculation from the register array or instruction parameters, and writes them back to the register array after calculation; the SALU reads the fields involved in the calculation from the register array or instruction parameters, and reads the data involved in the calculation from the data memory, and writes them back to the register array and / or the data memory after calculation; it is allowed to connect the matching module to the RTC processor or not. If the matching module is connected to the RTC processor, the current processor supports matching instructions, extracts the data required by the instruction from the register array and submits it to the matching module, and the matching module writes the matching result back to the register array.
[0018] Optionally, the reconfigurable processor allows being reconfigured into multiple RTC cores, and the number of RTC cores is determined according to the parallelism of the pipeline stage.
[0019] Optionally, it also includes: a parser, a payload buffer, a reverse parser and a hardware load, wherein the parser and the reverse parser are connected to the first side path, the payload buffer and the hardware load are connected to the second side path, and the parser is also connected to the payload buffer for parsing the data packet to obtain the PHV and the payload, and sending the payload to the payload buffer.
[0020] Optionally, the first side path includes multiple first-way nodes. In the first side path, after the PHV passes through the processing flow of the complete pipeline, it is sent to the corresponding first-way node by the last reconfigurable processor in the pipeline that is reconstructed as a pipeline stage. Among them, for the PHV that does not need to be sent to the RTC processor for processing, it is transmitted along the first path, and the first path is starting from the first-way node corresponding to the last reconfigurable processor in the pipeline that is reconstructed as a pipeline stage, and is directly transmitted to the first-way node corresponding to the inverse parser along the first side path; for the PHV that needs to be sent to the RTC processor for further processing, it is transmitted along the second path, and the second path is starting from the first-way node corresponding to the last reconfigurable processor in the pipeline that is reconstructed as a pipeline stage, and is transmitted along the first side path to the first-way node corresponding to the RTC processor where the RTC core assigned to the processing is located, and then downloaded to the RTC core assigned to the processing for processing. After the processing is completed, an idle cycle of the corresponding first-way node is selected and uploaded to the first-way node corresponding to the RTC processor where the RTC core that processes the PHV is located.
[0021] Optionally, the first path and the second path have a common source point and sink point, wherein the source point is the first path node corresponding to the last reconfigurable processor reconstructed as the pipeline stage, the sink point is the first path node corresponding to the inverse parser, and there is buffer processing congestion at the sink point. When the buffer meets the overflow condition, the buffer sends a target signal to the reconfigurable processor reconstructed as the RTC processor so as to no longer upload new PHVs to the second path based on the target signal.
[0022] Optionally, the second side path includes at least one second path node, and the second side path supports at least one data transfer, including: sending a request from a reconfigurable processor reconstructed as an RTC processor, reading the payload of a data packet processed by the reconfigurable processor reconstructed as the RTC processor in a payload buffer, and passing the payload back to the reconfigurable processor reconstructed as the RTC processor, wherein the payload will be stored in a data memory by the reconfigurable processor of the reconstructed RTC processor for subsequent processing; sending a request from a reconfigurable processor reconstructed as an RTC processor, reading data in a data memory or a register array on other reconfigurable processors, and passing the read data back to the data memory processed by the reconfigurable processor reconstructed as the RTC processor that sent the read request; sending a request from the reconfigurable processor reconstructed as the RTC processor, writing data to the data memory or register array on other reconfigurable processors; sending a request from the reconfigurable processor reconstructed as the RTC processor, sending data to a hardware load, and the hardware load sending the processing result back to the reconfigurable processor reconstructed as the RTC processor.
[0023] A second aspect of the present application provides a programmable data plane chip, comprising a processor-reconfigurable programmable switching structure of any one of the above embodiments.
[0024] Therefore, this application includes the following beneficial effects:
[0025] The embodiment of the present application significantly improves the programmability of the switch by reconfiguring the reconfigurable processor into a pipeline stage or RTC processor design while consuming only a small amount of chip area, so that network applications originally deployed on servers and middleboxes can be offloaded to programmable switches, reducing network communication delays, improving network service performance, and reducing the demand for servers and middlebox devices in data center networks, thereby reducing the construction and maintenance costs of data center networks. As a result, the problems of limited programmability of the programmable switching chip architecture, insufficient throughput, and high network communication delay in related technologies are solved.
[0026] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:
[0028] Figure 1 A schematic diagram of the Banzai architecture technology in the related art;
[0029] Figure 2 A schematic diagram of the Trio architecture technology in the related technology;
[0030] Figure 3 It is a schematic diagram of Sirius architecture technology in related technologies;
[0031] Figure 4 A block diagram of a programmable switching structure that provides a processor reconfiguration according to an embodiment of the present application;
[0032] Figure 5 An example diagram of the composition structure and data path of a reconfigurable processor provided according to an embodiment of the present application. DETAILED DESCRIPTION
[0033] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.
[0034] The following describes the processor-reconfigurable programmable switching structure and programmable data plane chip of the embodiment of the present application with reference to the accompanying drawings. In response to the problems mentioned in the above background technology, the present application provides a processor-reconfigurable programmable switching structure, which greatly improves the programmability of the switch by reconfiguring the reconfigurable processor into a pipeline stage or RTC processor design while consuming only a small amount of chip area, so that network applications originally deployed on servers and middleboxes can be offloaded to programmable switches, reducing network communication delays, improving the performance of network services, and reducing the demand for servers and middlebox devices in data center networks, thereby reducing the construction and maintenance costs of data center networks. As a result, the problems of limited architectural programmability of programmable switching chips, insufficient throughput, and high network communication delays in related technologies are solved.
[0035] Before introducing the specific content of this application, the relevant technical advances involved in this application are described in detail, including the following:
[0036] 1. Related Technology 1
[0037] like Figure 1 As shown in the figure, Banzai is the technical prototype of Intel's programmable switch Tofino. When a data packet enters, Banzai parses the header field of the data packet through a programmable parser, and processes the header through two pipelines at the inlet and outlet. These two pipelines with the same structure constitute Banzai's programmable packet processing logic. Each pipeline consists of a series of stages, namely MAU (Match Action Unit), which is the basic programmable unit of the prototype. Users program MAU in two aspects through the P4 program: what flow table the data packet needs to query at this stage, and what processing logic should be executed after the query results are obtained. In Banzai, the processing logic of each stage is performed using a set of parallel atomic operations, and each atomic operation can perform a "read-add / subtract-write back" function.
[0038] Programmable switching chips using pipeline architecture often have high throughput and low latency, but pipelines encounter difficulties when processing the following network tasks: (1) Large computational requirements. Since the processing logic in each MAU is performed using a set of parallel atomic operations, that is, only one cycle in each MAU actually performs computations, if the packet processing flow contains a large number of computations, the delay between two computations is a whole MAU, which seriously affects the packet processing delay. When the amount of computation reaches a certain level, the number of computations provided by the pipeline length is insufficient to accommodate the complete computation, so a recycle is required (that is, the data packet leaving the pipeline is resent to the pipeline head for processing again), resulting in a significant decrease in throughput. (2) Complex stateful functions. Due to the simplicity of atomic operations, only simple stateful functions (such as counters) can be implemented via atomic operations. For more complex functions, the logic for determining the next state may require a series of operations and table accesses, which cannot be implemented via atomic operations. In order to avoid bottlenecks in the pipeline, this technical solution cannot support these stateful functions by simply expanding the atomic operation circuit and enhancing the atomic operation function. Due to the unidirectional nature of the pipeline, there is no effective mechanism to write state data back to the early stage. The only solution is still re-circulation, which results in a significant reduction in throughput. (3) Long dependency chains between statements. Since the processing logic in each MAU is performed using a set of parallel atomic operations, if a statement in the packet processing flow needs to use the calculation result of another statement (i.e., it depends on this statement), these two statements cannot be executed in parallel and must be laid out in different MAUs. When the length of the dependency chain between statements reaches N, N different MAUs are required to execute this processing flow, even though a large amount of storage resources and computing resources in these N MAUs are wasted. This causes serious waste of resources, and for more complex packet processing flows (such as network service chains), the usual pipeline length (i.e., the number of MAUs) is not enough to lay out the complete flow, which requires the packet to be re-circulated, resulting in a significant reduction in throughput.
[0039] 2. Related Technology 2
[0040] Trio is the technical prototype of Juniper's MX series programmable switches. The unit that processes packets in Trio is PFE (Packet Forwarding Engine), and one PFE can be integrated on a chip. In order to expand the processing capacity of the switch, multiple PFEs need to be connected through a switching network. The structure of PFE is as follows: Figure 2As shown in the figure, it contains hundreds of PPEs (Packet Processing Engines). After the data packet enters the PFE, its payload is stored in the buffer, and the header is dispatched to the target PPE through a dispatcher and a fully connected network. Each PPE is an RTC processor, which can access the shared memory system (including SRAM (Static Random Access Memory) cache and off-chip DRAM (Dynamic Random Access Memory)) through the same fully connected network, hardware acceleration modules such as hashing and filtering, and the buffer for storing payloads. After the packet header is processed in the PPE, it is reordered by the reordering engine to ensure that the sending order is the same as the arrival order of the data packets, and then assembled with the payload in the buffer and sent out.
[0041] Because PPE is an RTC processor, the processing of a data packet can continue until the entire processing flow is completed, so Trio can execute longer and more complex processing flows without being limited by the length of the pipeline. However, it also faces the following challenges: (1) Insufficient throughput. Due to chip size and power limitations, the network throughput that a single PFE chip can support is quite limited, about 1-2 orders of magnitude lower than that of a programmable switching chip with a pipeline architecture. (2) High latency. PPE accesses memory (such as looking up flow tables and state tables) through a fully connected network from a shared storage system. The latency of a single memory access is about 1-2 orders of magnitude lower than that of a pipeline architecture programmable switching chip with MAU exclusive memory. For state tables where read and write conflicts may occur, synchronization mechanisms such as locking will further increase memory access latency. (3) State consistency is difficult to guarantee. Since the throughput of each PPE is very low, when processing large-scale stateful flows, the same flow may be dispatched to different PPEs, resulting in the order in which the data packets of the flow are processed being different from the order in which the data packets arrive. When the application requires strong state consistency or stricter bounded staleness consistency, the state consistency will be destroyed.
[0042] 3. Related Technology 3
[0043] Sirius is a technical prototype of Pensando DPU (Data Processing Unit) launched by AMD. Its architecture is as follows Figure 3As shown in the figure, it contains 2 ordinary pipelines and 3 dedicated pipelines for DMA (Direct Memory Access). Each pipeline in Sirius contains a TE (Table Engine) and several MPUs (Match Process Units). The function of the MPU is similar to that of the MAU in Banzai, but it does not contain memory and corresponding memory access modules. When the MPU needs to access memory, it accesses its shared storage system (including SRAM cache and off-chip DRAM) through the TE and a NOC (Network on Chip) interconnection network. The NOC interconnection network is also connected to the PCIe port leading to the server, hardware accelerators such as compression, encryption, CRC (Cyclic Redundancy Check), and a set of embedded ARM cores. When a data packet cannot complete the processing flow through the pipeline, it is sent to the ARM core working in the form of RTC for processing through the NOC interconnection network.
[0044] By integrating the pipeline and RTC architectures, Sirius achieves the separation of tasks with simple processing flows and tasks with complex processing flows. Simple tasks can be processed in a high-performance form through the pipeline, while complex tasks can be sent to the more programmable RTC core without being limited by the length of the pipeline. However, the existing fusion architecture still has the following problems: (1) It is not suitable for the switch scenario. The existing fusion architecture is designed for smart network cards, DPUs or IPUs (Infrastructure Processing Unit) scenarios. For example, the design bandwidth of AMD Pensando DSC2-200 is only 400Gbps, and this throughput cannot be used in the switch scenario. (2) Insufficient scalability. The fusion architecture always connects the pipeline and the RTC core through a fully connected bus or shared memory. These structures will increase in O(N) when the number of access modules N increases. 2) is rapidly expanding in complexity, resulting in unacceptable chip area and power consumption. (3) Idle resources. Due to the rapid development of software-defined networks, network switching equipment manufacturers are unable to accurately predict the network services that their network equipment will carry in the future, and different network services have significantly different requirements for pipelines and RTC cores. Services with stricter requirements on performance indicators such as throughput and latency may require more pipeline lengths, while services with complex processing flows may require more RTC cores. In order to meet user needs in different scenarios, network switching equipment manufacturers with converged architectures need to provide pipelines with the maximum length and the maximum number of RTC cores; however, when user services only require one type of computing resources, the other computing resources are idle, thereby reducing the effective utilization of chip area.
[0045] Specifically, Figure 4 A block diagram of a processor-reconfigurable programmable switching structure provided in an embodiment of the present application.
[0046] like Figure 4 As shown, the reconfigurable programmable switching structure of the processor includes: a first side path, a second side path and a reconfigurable processor sequence.
[0047] The first side path is used to transmit a data packet header vector PHV, where PHV is a signal composed of fields to be processed in the data packet header and other metadata fields generated during the data packet processing, arranged in a certain order, and its typical bit width is 4096 bits; the second side path is used to transmit other data and control signals other than PHV; a reconfigurable processor sequence, the reconfigurable processor sequence includes multiple reconfigurable processors, the first side path and the second side path connect the multiple reconfigurable processors, wherein each reconfigurable processor of the multiple reconfigurable processors is allowed to be reconfigured into a pipeline stage or an RTC processor, and RTC is a data packet running the entire processing flow on one processor until completion, and will not be sent to other processors, corresponding to the process in which each stage in the pipeline structure only processes one link, and there is a data path for transmitting PHV between each adjacent reconfigurable processor, when both ends of the data path are reconfigurable processors reconfigured into pipeline stages, the data path is opened to connect the two pipeline stages into a complete pipeline; when both ends of the data path are not reconfigurable processors reconfigured into pipeline stages, the data path is closed, and the reconfigurable processors reconfigured into RTC processors are connected through the first side path.
[0048] In the embodiment of the present application, the reconfigurable processor is a special processor structure, which can be reconfigured into a stage of the pipeline (for example, a MAU) or an RTC processor according to the content of the configuration register sent by the compiler. Figure 5 As shown, it contains a series of hardware components and the connections between the hardware components. When it is reconstructed into a pipeline or RTC architecture, specific components and connections between components are made effective, thereby realizing the reuse of hardware resources.
[0049] Furthermore, the embodiments of the present application adopt Figure 4 The structure is used to connect various reconfigurable processors in series, which mainly includes a group of linearly arranged reconfigurable processor sequences, a first side path for transmitting PHV and a second side path for transmitting other data and control signals.
[0050] Among them, after the data packet passes through the parser, the parsed PHV enters the first reconfigurable processor in the reconfigurable processor sequence (if all reconfigurable processors are reconfigured into RTC processors, they enter the first node of the first side path), and the remaining loads enter the buffer SRAM for storage. There is a data path between the reconfigurable processors that can transmit PHV. When both ends of the data path are reconfigurable processors reconfigured into pipeline stages, the data path is opened to connect the two pipeline stages to a complete pipeline; when both ends of the path are not reconfigurable processors reconfigured into pipeline stages, the data path is closed and PHV is not transmitted. Therefore, in the entire reconfigurable processor sequence, the first several (can be 0) reconfigurable processors are reconfigured into pipeline stages and connected into a complete pipeline through the data path between the reconfigurable processors. The latter several (can be 0) reconfigurable processors are reconfigured into RTC processors, each RTC processor is relatively independent and is only connected through the first side path.
[0051] It should be noted that the embodiment of the present application reconstructs the parallelism of the processor when it is reconstructed into a pipeline stage, and determines the number of RTC cores when it is reconstructed into an RTC processor. When the reconfigurable processor uses a larger chip area, it has a greater degree of parallelism when it is used as a pipeline stage (such as being able to search more tables at the same time), and it has more cores when it is used as an RTC processor. This makes the embodiment of the present application have good scalability. When the manufacturer is willing to pay more chip area and power consumption, the performance can be improved linearly (under the limitation of hardware wiring capabilities). In addition, side pathways include but are not limited to ring interconnections, interconnections with similar principles such as unidirectional or bidirectional linear interconnections, and ring interconnections such as Figure 4 As shown, the side passage of the embodiment of the present application is Figure 4 The example shown is a ring interconnection.
[0052] In one embodiment of the present application, the reconfigurable processor includes a processor in which all of the following items are reconfigurable, or a processor in which some of the following items are reconfigurable and other items are not included in the processor, or a processor in which some of the following items are reconfigurable and other items are not reconfigurable, and also includes a register array, a data memory, an instruction memory, an ALU, a SALU, and a matching module. Each reconfigurable processor allows being reconfigured into a pipeline stage or an RTC processor. The reusable components between the pipeline stage and the RTC processor in the embodiment of the present application are as follows:
[0053] 1) Register array. In the pipeline stage, the register array is used to store data packet header fields, metadata, and intermediate signals used in the processing flow. In order to avoid pipeline blocking, these data are required to be stored in the register array instead of RAM. In the RTC processor, the register array is used to store general and special register groups.
[0054] 2) Data memory. In the pipeline stage, the data memory is used to store the lookup table for data packets to look up when they pass through, and to perform corresponding operations based on the lookup results; the data memory is also used to store the state table for reading, modifying, and writing back the state when data packets pass through. In the RTC processor, the data memory is used to store data packet header fields, metadata, lookup tables, and state tables, as well as intermediate signals used in the processing flow. If the RTC processor is equipped with an operating system, the data memory is also used to store data structures such as stacks maintained by the operating system.
[0055] 3) Instruction memory: It is used to store instructions in both the pipeline stage and the RTC processor, but the instruction format is different.
[0056] 4) ALU. Used to perform arithmetic logic calculations in both the pipeline stage and the RTC processor.
[0057] 5) SALU. In the pipeline stage, it is used to read the state from the state table of the data memory, perform arithmetic and logical calculations, and write the calculation results back to the state table. In the RTC processor, it is used to read data from the data memory (not limited to the state table), perform arithmetic and logical calculations, and write the calculation results back to the data memory.
[0058] 6) Matching module: In both the pipeline stage and the RTC processor, it is used to search and match in the lookup table of the data memory according to the given keyword, and return whether there is a matching table entry and the corresponding table entry content.
[0059] It should be noted that in order to implement a reconfigurable processor, the embodiment of the present application reuses the above-mentioned multiple hardware components. Reusing any one of the hardware components to implement the reconfigurable technology is not limited to reusing each reusable component mentioned in the embodiment of the present application.
[0060] In one embodiment of the present application, when the reconfigurable processor is reconfigured into a pipeline stage, the parsed data packet header and the metadata and other fields (i.e., PHV) carried by it are received by the current processor and written into the register array of the current processor. According to the keyword selection rule specified by the user program, a part of the fields are selected to form the keyword and transmitted to the matching module. The matching module searches and matches in the lookup table of the data storage according to the given keyword. When there is a matching table entry, the operation ID to be executed is determined according to the content of the table entry; when there is no matching table entry, the default operation ID specified by the user program is used. According to the operation ID, the corresponding instruction is read from the instruction memory, the instruction is parsed using a dedicated decoder, and it is dispatched to the ALU and SALU. The ALU reads the fields involved in the calculation from the register array or instruction parameters, and writes them back to the register array after calculation. The SALU reads the fields involved in the calculation from the register array or instruction parameters, and reads the state involved in the calculation from the state table in the data storage, and writes them back to the register array and / or the state table in the data storage after calculation.
[0061] In the above-mentioned whole processing flow, the PHV of a data packet is transferred in the register array corresponding to the registers of different pipeline beats to realize the uninterrupted pipeline function. After the processing is completed, the corresponding PHV is sent out of the current processor.
[0062] In one embodiment of the present application, when the reconfigurable processor is reconfigured into the RTC processor, after the PHV is received by the current processor, it is written into the data memory of the current processor and notifies the current processor to start executing the processing flow. The RTC processor continuously reads the instructions to be executed from the instruction register according to the PC (Program Counter) until the processing termination instruction is read.
[0063] Each instruction read is parsed by a dedicated decoder and dispatched to the ALU or SALU. The ALU reads the fields involved in the calculation from the register array or instruction parameters, and writes them back to the register array after calculation. The SALU reads the fields involved in the calculation from the register array or instruction parameters, and reads the data involved in the calculation from the data memory, and writes them back to the register array and / or data memory after calculation. The branch jump instruction in the instruction will change the content of the PC.
[0064] The RTC processor can write the extracted key code into the register array and submit it to the matching module through a specific instruction. The matching module works as a coprocessor and writes the result back to the register array. After reading the processing termination instruction, the corresponding PHV is sent out of the current processor.
[0065] It should be noted that a reconfigurable processor can be reconfigured into multiple RTC cores, and the number of convertible RTC cores is determined by the parallelism of the pipeline stage. In the pipeline stage, instructions may be given in the form of VLIW (Very Long Instruction Word), and are parsed and dispatched to multiple ALUs or SALUs for parallel operation. At the same time, the pipeline stage will provide multiple memory access channels and multiple functional accelerators (hashing, etc., usually integrated in the matching module). Therefore, multiple groups of parallel functional components can be assigned to different cores in the RTC processor, thereby realizing the reconstruction of a pipeline stage into multiple RTC cores.
[0066] The RTC processor does not participate in the process of receiving and sending PHVs, and can process other PHVs while receiving and sending PHVs. Each RTC core contains a PHV buffer to temporarily store PHVs that have been received but not yet processed, and PHVs that have been processed but not yet sent.
[0067] It should be noted that the above content is the part of the reconfigurable processor that is effective when the reconfigurable processor is reconfigured into a pipeline stage and when the reconfigurable processor is reconfigured into an RTC processor, rather than the entire part of the reconfigurable processor. The entire part of the reconfigurable processor also includes:
[0068] 1. An instruction decoder dedicated to the pipeline stage and an instruction decoder dedicated to the RTC processor;
[0069] 2. Configuration registers dedicated to the pipeline stages and configuration registers dedicated to the RTC processor and other registers (including the program counter);
[0070] 3. A circuit that connects all the above modules in a specific way;
[0071] 4. For the configuration register of the above circuit, according to the state of the configuration register, part of the circuit is connected and the other part is disconnected. The circuit connected to the switching structure can enable the reconfigurable processor to be connected to the entire switching structure as a pipeline stage or RTC processor.
[0072] Furthermore, the reconfigurable programmable switching structure of the processor also includes: a parser, a load buffer, a reverse parser and a hardware load.
[0073] Specifically, the parser, the inverse parser and each reconfigurable processor are connected to the first side path and hold a corresponding external path node. The first side path includes multiple first path nodes. After the PHV passes through the complete pipeline processing flow, it is sent to its corresponding first path node by the last reconfigurable processor reconfigured as a pipeline stage in the pipeline. The node determines whether the PHV needs to be sent to the RTC processor for further processing and which RTC processor it needs to be sent to for further processing according to a dedicated field in the PHV.
[0074] For the PHV that does not need to be sent to the RTC processor for further processing, it starts from the first-way node corresponding to the last reconfigurable processor in the pipeline that is reconstructed into a pipeline stage, and is directly transmitted along the first side path to the first-way node corresponding to the inverse parser, which is the first path.
[0075] For the PHV that needs to be sent to the RTC processor for further processing, the first node corresponding to the last reconfigurable processor reconstructed as a pipeline stage in the pipeline is started, and is transmitted to the corresponding RTC processor along the first side path, and downloaded to the RTC core assigned to it for processing. After the processing is completed, the idle cycle of the corresponding first node is selected, sent to the first node and continued to be transmitted forward to the first node corresponding to the inverse parser, which is the second path.
[0076] Furthermore, the first path and the second path in the embodiment of the present application have a common source point and sink point, the source point is the first path node corresponding to the last reconfigurable processor reconfigured as a pipeline stage, the sink point is the first path node corresponding to the inverse parser, and other nodes are not shared, so as to ensure that data packets passing through the first path that do not require complex processing can always be processed with low latency. When a reconfigurable processor is reconfigured into an RTC processor containing more than one core, a scheduling module is included at the first path node corresponding to each reconfigurable processor, and each RTC core on the reconfigurable processor is scheduled to be sent to the PHV on the first path node using the Round-Robin method.
[0077] In addition, the first path and the second path of the embodiment of the present application have a common confluence point, that is, the first path node corresponding to the inverse parser. There is a buffer at the confluence point to handle temporary congestion. When this buffer is about to overflow, it sends a signal to the reconfigurable processor reconstructed as the RTC processor, requiring it to no longer upload new PHVs to the second path. This mechanism allows the data memory used to store PHVs in all RTC processors to be used as a buffer for handling congestion at the confluence point. Every time a data packet enters this buffer, it means that the first path is idle within a cycle, so that a data packet of the second path can be scheduled, thereby ensuring that the number of data packets on the second path does not increase when this mechanism is triggered, thereby ensuring that the congestion at the confluence point can be resolved without packet loss.
[0078] In one embodiment of the present application, each reconfigurable processor, load buffer, and other hardware loads (such as shared memory and accelerator, etc.) are connected to the second side path and have a corresponding second path node. The second side path includes at least one second path node, and the second side path is used to control the transmission of signals and data, and can support at least one of the following data transmissions:
[0079] 1) Send a request from the reconfigurable processor reconstructed as the RTC processor, read the payload of the data packet processed by the reconfigurable processor reconstructed as the RTC processor in the payload buffer, and pass the payload back to the reconfigurable processor reconstructed as the RTC processor to implement DPI (Deep Packet Inspection), wherein the payload will be stored in the data memory by the reconfigurable processor of the reconstructed RTC processor for subsequent processing.
[0080] 2) Send a request from the reconfigurable processor reconstructed as the RTC processor to read data in the data memory or register array on other reconfigurable processors, and pass the read data back to the data memory processed by the reconfigurable processor reconstructed as the RTC processor that sent the read request.
[0081] 3) Send a request from the reconfigurable processor reconfigured as the RTC processor to write data to the data memory or register array on the other reconfigurable processors.
[0082] 4) Send a request from the reconfigurable processor reconfigured as the RTC processor, send data to the hardware load, and the hardware load sends the processing result back to the reconfigurable processor reconfigured as the RTC processor.
[0083] The following embodiments of the present application can be illustrated by taking the in-network computing applications that the current chip cannot support but the present application can support as examples, including the following application scenarios:
[0084] Application scenario 1 (intra-network aggregation):
[0085] As large model technology stands out in various artificial intelligence application scenarios, distributed neural networks are widely deployed in data center networks. The computing nodes in the network need to perform a large amount of communication to synchronize training parameters. One of the important operations is to aggregate the gradients calculated by each computing node. The traditional approach is to use a server for aggregation, which will cause a large communication burden on the server network interface, thereby reducing the convergence speed of the entire neural network training. ATP proposes a solution for in-network aggregation on a programmable switch, but due to the pipeline length limitation of the current chip, the in-network aggregation capability of this solution on the programmable switch is limited. At the same time, the current chip cannot achieve effective functional isolation when used for in-network aggregation. When processing in-network aggregated traffic, the ability of the programmable switch to process non-in-network aggregated traffic will be significantly reduced, so that other locally available bandwidth is limited, and the overall service quality of the cloud data center network is reduced. The use of Trio can break the pipeline length limitation, but still cannot achieve functional isolation, and the in-network aggregated traffic will still significantly reduce the overall performance of the switch. The present invention can break the limitation of the pipeline length on the aggregation capability by sending the in-network aggregated traffic to the RTC core for processing, and effectively isolate it from other traffic processed on the pipeline.
[0086] Application scenario 2 (on-network cache):
[0087] In-network cache technology is very effective for accelerating query server cache. A portion of the server query results are cached on the switch. If the newly arrived query data packet hits the cache, the query results in the cache are directly returned without forwarding to the server for query; if the cache is not hit, the server returns the query results through the switch, and the corresponding query results are written into the cache to replace the outdated entries. NetCache uses a controller to handle the cache replacement function, which causes a performance bottleneck. P4LRU implements an LRU cache replacement function on the pipeline, but due to the pipeline programmability limitations of existing chips, each group of LRU can only support 3 entries, so that when the data locality is not particularly good, the hit rate of this method is much lower than that of the ideal LRU. At the same time, under the existing technology, the memory resources on the pipeline cannot be fully utilized to store in-network cache entries, so that the in-network cache cannot fully exert the memory potential on the switch. The present invention uses the pipeline as a level 1 cache, and forwards it to the RTC core to search for the level 2 cache under the condition that the level 1 cache does not hit. With the RTC core's ability to maintain complex data structures, the second-level cache on each core can implement a better cache replacement algorithm; at the same time, the RTC core can use all available memory on the processor as cache, so that under the same scale of resources, more in-network cache entries can be stored on the switch; ultimately, the in-network cache hit rate is significantly improved, further reducing the network load and computing load on the server side.
[0088] Application scenario 3 (network service integration):
[0089] In data center networks and edge networks, specialized middleboxes and accelerators that support various network functions are usually deployed to enhance network security and performance. For example, gateway switches require DDoS detection and mitigation (such as SYN flood detection). Other basic network functions include server load balancing and network address translation. In addition, telemetry and measurement (such as heavy hit detection and in-band network telemetry) designed to improve network visibility are becoming indispensable for improving efficiency and ensuring service quality. Deploying multiple single-purpose middleboxes is not only costly, but also complicates network configuration and management. Integrating certain network functions into programmable data plane devices helps reduce costs. On existing pipeline-based chips, due to the complexity of network functions, it may be necessary to circulate data packets multiple times, resulting in huge losses in throughput and latency. On existing multi-core RTC architecture chips, the processing performance of the chip is also greatly reduced due to complex network functions. The present invention realizes an organic combination of the pipeline and multi-core RTC processors by dividing network services into the pipeline and the multi-core RTC processor. The basic and common functions applicable to all traffic are processed by the pipeline, while the complex functions applicable only to selected data packets are processed by the RTC core. Thanks to the reconfigurable processor design of the present invention, even without prior knowledge of applications and traffic distribution, the way network services are divided on pipelines and multi-core RTC processors can be changed through reconstruction, so that the overall processing performance is optimized.
[0090] According to the processor-reconfigurable programmable switching structure proposed in the embodiment of the present application, by reconfiguring the reconfigurable processor into a pipeline stage or RTC processor design, the programmability of the switch is greatly improved while consuming only a small amount of chip area, so that network applications originally deployed on servers and middleboxes can be offloaded to programmable switches, reducing network communication delays, improving network service performance, and reducing the demand for servers and middlebox devices in data center networks, thereby reducing the construction and maintenance costs of data center networks. As a result, the problems of limited architectural programmability, insufficient throughput, and high network communication delays in programmable switching chips in related technologies are solved.
[0091] An embodiment of the present application also provides a programmable data plane chip, including a programmable switching structure in which the processor of the above embodiment can be reconfigured.
[0092] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.
[0093] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
[0094] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.
[0095] It should be understood that the various parts of the present application can be implemented in hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, the steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array, a field programmable gate array, etc.
[0096] A person of ordinary skill in the art may understand that all or part of the steps carried by the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the above-mentioned program may be stored in a computer-readable storage medium, which, when executed, includes one of the steps of the method embodiment or a combination thereof.
[0097] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.
Claims
1. A processor-reconfigurable programmable switching structure, characterized in that: include: A first side channel for transmitting a data packet header vector PHV; A second side channel is used to transmit other data and control signals other than the PHV; A reconfigurable processor sequence, the reconfigurable processor sequence comprising a plurality of reconfigurable processors, the first side path and the second side path connecting the plurality of reconfigurable processors, wherein each of the plurality of reconfigurable processors is allowed to be reconfigured into a pipeline stage or run to complete an RTC processor, a data path for transmitting the PHV exists between each adjacent reconfigurable processor, when both ends of the data path are reconfigurable processors reconfigured into pipeline stages, the data path is opened to connect two pipeline stages into a complete pipeline, and when both ends of the data path are not reconfigurable processors reconfigured into pipeline stages, the data path is closed, and the reconfigurable processors reconfigured into RTC processors are connected via the first side path.
2. The processor-reconfigurable programmable switching structure according to claim 1, characterized in that: The reconfigurable processor includes a processor in which all of the following items are reconfigurable, or a processor in which some of the following items are reconfigurable and the other items are not included in the processor, or a processor in which some of the following items are reconfigurable and the other items are not reconfigurable: A register array, in the pipeline stage, the register array is used to store data packet header fields, metadata and intermediate signals, and in the RTC processor, the register array serves as a register group specified by the instruction system; Data storage, in the pipeline stage, the data storage is used to store lookup tables and state tables, the lookup tables are used for lookups when data packets pass through, and the state tables are used for reading, modifying, and writing back states when data packets pass through. In the RTC processor, the data storage is used to store data packet header fields, metadata, lookup tables and state tables, and intermediate signals. If the RTC processor is equipped with an operating system, the data storage is also used to store data structures related to operating system maintenance; Instruction memory, used to store instructions in both the pipeline stage and the RTC processor; Arithmetic logic unit ALU, used to perform arithmetic logic calculations in both the pipeline stage and the RTC processor; The stateful arithmetic logic unit SALU is used to read the state from the state table of the data memory in the pipeline stage, perform arithmetic logic calculations, and write the calculation results back to the state table; In the RTC processor, it is used to read data from the data memory, perform arithmetic and logical calculations, and write the calculation results back to the data memory; The matching module is used in both the pipeline stage and the RTC processor to search and match in the lookup table of the data memory according to the given keyword, and return whether there is a matching table entry and the corresponding table entry content.
3. The processor-reconfigurable programmable switching structure according to claim 2, characterized in that: When the reconfigurable processor is reconfigured into a pipeline stage, it includes: After the PHV of the data packet is received by the current processor, it is written into the register array of the current processor; according to the keyword selection rule specified by the user program, a part of the fields is selected to form the keyword and transmitted to the matching module; The matching module searches and matches in the lookup table of the data storage according to the given keywords. When there is a matching table entry, the execution operation ID is determined according to the table entry content; when there is no matching table entry, the default operation ID specified by the user program is used; Read instructions from the instruction memory according to the operation ID, parse the instructions using the decoder, and dispatch the parsed instructions to the ALU and SALU, wherein the ALU reads the fields involved in the calculation from the register array or the instruction parameter, and writes them back to the register array after calculation; the SALU reads the fields involved in the calculation from the register array or the instruction parameter, and reads the states involved in the calculation from the state table in the data memory, and writes them back to the register array and / or the state table in the data memory after calculation; In the processing flow, the PHV of a data packet is transferred between registers of the register array corresponding to different pipeline beats, and after the processing is completed, the corresponding PHV is sent out of the current processor.
4. The processor-reconfigurable programmable switching structure according to claim 2, characterized in that: When the reconfigurable processor is reconfigured into an RTC processor, it includes: After the PHV of the data packet is received by the current processor, it is written into the data memory of the current processor and notifies the current processor to start executing the processing flow. The current processor reads the instructions to be executed from the instruction register according to the program counter until the processing termination instruction is read; Each instruction read is parsed by a dedicated decoder and dispatched to the ALU or SALU, wherein the ALU reads the fields involved in the calculation from the register array or instruction parameters, and writes them back to the register array after calculation; the SALU reads the fields involved in the calculation from the register array or instruction parameters, and reads the data involved in the calculation from the data memory, and writes them back to the register array and / or data memory after calculation; The matching module is allowed to be connected to or not connected to the RTC processor. If the matching module is connected to the RTC processor, the current processor supports matching instructions, extracts data required by the instructions from the register array and submits it to the matching module, and the matching module writes the matching results back to the register array.
5. The processor reconfigurable programmable switching structure according to any one of claims 1 to 4, characterized in that: The reconfigurable processor allows being reconfigured into a plurality of RTC cores, and the number of RTC cores is determined according to the parallelism of the pipeline stage.
6. The processor reconfigurable programmable switching structure according to claim 1, characterized in that: Also includes: A parser, a payload buffer, a reverse parser and a hardware payload, wherein the parser and the reverse parser are connected to the first side path, the payload buffer and the hardware payload are connected to the second side path, and the parser is also connected to the payload buffer for parsing data packets to obtain PHVs and payloads, and sending the payloads to the payload buffer.
7. The processor-reconfigurable programmable switching structure according to claim 6, characterized in that: The first side path includes a plurality of first-way nodes, in which the PHV is sent to the corresponding first-way node by the last reconfigurable processor in the pipeline that is reconfigured as a pipeline stage after the PHV passes through the processing flow of the complete pipeline, wherein: For PHVs that do not need to be sent to the RTC processor for processing, they are transmitted along the first path, where the first path starts from the first way node corresponding to the last reconfigurable processor in the pipeline that is reconfigured as a pipeline stage, and is directly transmitted to the first way node corresponding to the inverse resolver along the first side path; For the PHV that needs to be sent to the RTC processor for further processing, it is transmitted along the second path, which starts from the first node corresponding to the last reconfigurable processor in the pipeline that is reconstructed into a pipeline stage, and is transmitted along the first side path to the first node corresponding to the RTC processor where the RTC core assigned to it for processing is located, and downloaded to the RTC core assigned to it for processing. After the processing is completed, the idle cycle of the corresponding first node is selected and uploaded to the first node corresponding to the RTC processor where the RTC core that processes the PHV is located.
8. The processor-reconfigurable programmable switching structure according to claim 7, characterized in that: The first path and the second path have a common source point and a sink point, wherein the source point is a first-way node corresponding to the last reconfigurable processor reconstructed as a pipeline stage, the sink point is a first-way node corresponding to the inverse parser, and a buffer is processed at the sink point. When the buffer meets an overflow condition, the buffer sends a target signal to the reconfigurable processor reconstructed as an RTC processor, so as not to upload a new PHV to the second path based on the target signal.
9. The processor-reconfigurable programmable switching structure according to claim 6, characterized in that: The second side path includes at least one second path node, and the second side path supports at least one data transmission, including: Sending a request from a reconfigurable processor reconfigured as an RTC processor, reading a payload of a data packet processed by the reconfigurable processor reconfigured as the RTC processor in a payload buffer, and transferring the payload back to the reconfigurable processor reconfigured as the RTC processor, wherein the payload will be stored in a data memory by the reconfigurable processor of the reconfigured RTC processor for subsequent processing; Sending a request from the reconfigurable processor reconfigured as the RTC processor to read data in a data memory or a register array on other reconfigurable processors, and transferring the read data back to the data memory processed by the reconfigurable processor reconfigured as the RTC processor; Sending a request from the reconfigurable processor reconfigured as the RTC processor to write data to the data memory or register array on the other reconfigurable processors; A request is sent from the reconfigurable processor reconfigured as the RTC processor, data is sent to a hardware load, and the hardware load sends a processing result back to the reconfigurable processor reconfigured as the RTC processor.
10. A programmable data plane chip, characterized in that: A programmable switching structure with processor reconfigurability comprising the one described in any one of claims 1-9.
Citation Information
Patent Citations
Easily programmable memory controller which can access different speed memory devices on different cycles
CA2160200A1
Reconfigurable processor chip driven by data stream and reconfigurable processor cluster
CN116303225A
Reconfigurable packet protocol parser equipment of network switching chip with hundred gigabit rate
CN117880395A
Prediction execution-based band state programmable data plane structure and chip
CN118259887A
Efficient conditional instruction having companion load predicate bits instruction
US20150370562A1
Cited By
Reconfigurable processor-based programmable switching fabric and programmable data plane chip
US12425356B1
