Communication management system and equipment for DPU (Data Processing Unit) card and onboard controller
By integrating the onboard controller on the DPU card and using multiple interfaces for communication management, the problem of inconvenient DPU card resource management in the data center is solved, efficient system management and remote monitoring are achieved, and system performance and flexibility are improved.
Patent Information
- Application Number
- CN202411229368.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-03
- Publication Date
- 2024-12-20
AI Technical Summary
The existing technology is difficult to effectively manage and optimize the system and hardware resources of DPU cards in data centers, resulting in the imbalance in the development of computing power and communication infrastructure.
Design a communication management system for DPU card and on-board controller. Through the integrated on-board controller, the communication management is carried out using I2C, SPI, PECI, IPMI and Redfish interfaces to realize the configuration of CPU processor and external sensors. , collect, monitor, and provide remote management functions.
It improves the management convenience and flexibility of DPU cards, reduces system complexity and power consumption, forms a compact system-level chip, supports customized requirements and remote monitoring management, and improves system performance and efficiency.
Smart Images
Figure CN119166460A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present application relate to the field of network communication technology, and in particular to a DPU card and onboard controller communication management system and equipment. Background Art
[0002] With the rapid development of data centers, communication capabilities and computing capabilities have become two important development directions of data center infrastructure. In order to cope with the exponential growth of data volume and complexity, DPU came into being. The background of its proposal is to cope with the problem of imbalance between the improvement of computing power and the development of communication infrastructure in data centers. DPU is a processor specially used for data processing. Unlike general-purpose processors (CPUs) and graphics processing units (GPUs), it focuses on data processing to achieve more efficient data processing and calculation. The current invention is that in modern computer systems, the data processing unit (DPU) card is a key component that is responsible for processing large amounts of data and performing fast calculations. With the growing demand for cloud computing and big data processing, the requirements for the performance and efficiency of DPU cards are also getting higher and higher. There is an urgent need to design a software or hardware solution to manage the system and software and hardware resources of the DPU card. Summary of the invention
[0003] The embodiment of the present application provides a DPU card and onboard controller communication management system, which improves the convenience and flexibility of DPU card management.
[0004] In a first aspect, an embodiment of the present application provides a DPU card and onboard controller communication management system, including:
[0005] An onboard controller is integrated on the DPU card, and the onboard controller is used to manage and control the configuration, collection and monitoring of the CPU processor and external sensors on the mainboard;
[0006] The DPU card and the onboard controller communicate via I2C or SPI bus.
[0007] In a possible implementation manner, the DPU card and the onboard controller are managed and controlled through at least one of a PECI interface, an IPMI interface, and a Redfish interface.
[0008] In a possible implementation, the onboard controller obtains the temperature information and power management information of the CPU through the PECI interface or the IPMI interface.
[0009] In a possible implementation, the onboard controller remotely monitors and manages the hardware status of the server through the IPMI interface.
[0010] In a possible implementation, the onboard controller remotely manages an operation interface through the Redfish interface, and the operation interface is used to query, configure, modify or upgrade the DPU card.
[0011] In a possible implementation manner, the target DPU card is connected via a PCIE bus, and the DPU card is used to manage at least one of the target DPU cards.
[0012] In a possible implementation, the onboard controller is also used to provide log management, user management, BIOS management, or alarm management.
[0013] In a second aspect, an embodiment of the present application further provides an electronic device, the electronic device comprising:
[0014] at least one processor;
[0015] and a memory communicatively coupled to the at least one processor;
[0016] Among them, the memory stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor to enable the electronic device to execute a management system corresponding to any embodiment in the first aspect of the embodiments of the present application.
[0017] In summary, in the system shown in the embodiment of the present application, the onboard controller is highly integrated with other components of the DPU to form a compact system-on-chip (SoC), which improves system performance and reduces power consumption. Customized requirements can be provided according to scenarios and needs. Remote management functions can also be provided, allowing users to remotely monitor and manage servers through the network, improving the convenience and flexibility of management. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.
[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0020] Figure 1 A schematic diagram of the interaction between the CPU and the onboard controller provided in the embodiment of the present application;
[0021] Figure 2 The DPU card and onboard controller communication management system provided in the embodiment of the present application;
[0022] Figure 3 An electronic device for a DPU card and onboard controller communication management system provided in an embodiment of the present application. DETAILED DESCRIPTION
[0023] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments are only examples of systems consistent with some aspects of the embodiments of the present application as detailed in the appended claims.
[0024] The following specific embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments. The following will describe the embodiments of the present application in conjunction with the accompanying drawings.
[0025] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0026] Figure 1 A schematic diagram of the interaction between the CPU and the onboard controller provided in the embodiment of the present application. Figure 1 The communication forms and interfaces that may be used between the baseboard controller (Baseboard Management Controller) and the CPU are included. The BMC chip and the CPU chip are on the same motherboard, and the connection between the CPU and the BMC chip mainly involves data communication and the transmission of control signals. Although the specific connection method and protocol may vary depending on the design of the DPU and BMC, the following are some common connection methods and features: PECI (Platform Environment Control Interface) interface, PECI is a one-wire bus used to monitor the temperature of the CPU and chipset. The BMC may obtain the temperature information of the CPU through the PECI interface and adjust the fan speed and other controls based on this information. The PECI method reads the average temperature within a 256ms time window, and an independent communication bus can be used in the CPU's C0 to C6 states.
[0027] In addition to the PECI interface, the CPU and BMC can also use other independent communication buses for data transmission and control signal exchange. These buses may include I2C (Inter-Integrated Circuit), SPI (Serial Peripheral Interface) or other customized communication protocols. Power management: Since the DPU needs to continue to work when the server is shut down, the BMC may be involved in the power management of the DPU. This may include that when the server enters the S5 state, the BMC can recognize that the DPU smart network card is in place and turn on all the power out-of-band management networks required by the DPU.
[0028] The BMC in the DPU may be connected through an out-of-band management network to enhance the security, reliability, and availability of the system. This network is independent of the main data network and is used for remote monitoring, configuration, and management of the DPU. Integration In some DPU designs, the BMC is highly integrated with other components of the DPU to form a compact system-on-chip (SoC). This integration can improve system performance, reduce power consumption, and complexity customization.
[0029] Since DPU has a wide range of application scenarios, including ultra-large-scale HPC / AI, cloud computing and storage applications, the connection between the CPU and BMC can also be customized according to specific application requirements. When the DPU is equipped with BMC, the connection between the CPU and the BMC chip is a complex system involving multiple communication interfaces, power management, and out-of-band management networks. The choice of these connection methods and protocols depends on the specific design, application scenarios, and performance requirements of the DPU. The communication protocol between the CPU and the BMC (baseboard management controller) chip depends on the specific server design and manufacturer implementation. This application embodiment does not limit this.
[0030] It is understandable that although I2C is mainly used for low-speed, short-distance communication, in some designs it can be used as one of the communication protocols between the CPU and the BMC.
[0031] Figure 2 A DPU card and onboard controller communication management system is provided for the embodiment of the present application. Figure 2 The process includes S101 and S102. The details are as follows.
[0032] S101. Integrate an onboard controller BMC on the DPU card. The onboard controller is used to manage and control the configuration, collection and monitoring of the CPU processor and external sensors on the mainboard.
[0033] In the embodiment of the present application, the BMC is integrated on the DPU card, which can reduce the additional circuit design and layout space, thereby reducing the design complexity and cost of the DPU card. Since the BMC is integrated on the DPU card, it does not need to communicate with other components of the DPU card through a specific interface, so it can avoid introducing additional delays and overhead, thereby improving the performance and efficiency of the server.
[0034] S102: The DPU card and the onboard controller communicate via an I2C or SPI bus.
[0035] In a possible implementation, the DPU card and the onboard controller are managed and controlled through at least one of a PECI interface, an IPMI interface, and a Red fish interface.
[0036] It is understandable that the DPU card and the onboard controller can be managed and controlled through the PECI interface, can be managed and controlled through the IPMI interface, and can also be managed and controlled through the Redfish interface. This embodiment of the application does not limit this.
[0037] In a possible implementation, the onboard controller obtains the temperature information and power management information of the CPU through the PECI interface or the IPMI interface.
[0038] In a possible implementation, the onboard controller remotely monitors and manages the hardware status of the server through the IPMI interface.
[0039] In the embodiment of the present application, IPMI (Intelligent Platform Management Interface) is a management interface independent of the operating system, which allows the server host to communicate with the BMC via Ethernet. IPMI provides a series of protocols and commands for remote monitoring and management of the server's hardware status, such as CPU temperature, fan speed, power status, etc. Administrators can use IPMI to remotely restart the server, update firmware, collect logs, etc. IPMI communication usually uses IPMI-over-LAN or IPMI-over-Serial or middle layer software protocol (ME).
[0040] It can be seen that the BMC provided in the embodiment of the present application can provide a remote management function, allowing a user to remotely monitor and manage the server through a network, thereby improving the convenience and flexibility of management.
[0041] In some designs, the CPU and BMC may communicate through an intermediate software protocol (such as ME). These protocols allow hardware-specific interactions between the BMC and the CPU, such as the I2C (Inter-Integrated Circuit) protocol, which is a bidirectional two-wire synchronous serial bus used to connect microcontrollers and their functional peripherals.
[0042] In a possible implementation, the onboard controller performs remote management of the operation interface through the Red Fish interface as an operation interface, and the operation interface is used to query, configure, modify or upgrade the DPU card.
[0043] In the embodiment of the present application, BMC uses IPMI protocol to provide access interface to solve the monitoring problem of heterogeneous server motherboards. Redfish is the next generation of IPMI. As an open industry standard specification, it provides simple, modern and secure management functions for scalable platform hardware. The unified management interface specification based on Redfish can effectively reduce the large amount of adaptation, development and testing work brought by different server hardware management interfaces. At the same time, the onboard management control system (BMC), as an important part of server hardware management, is responsible for monitoring and managing various components of the server, including CPU processors and external sensors.
[0044] It can be seen that the BMC provided in the embodiment of the present application can not only manage the hardware components of the server, but also manage the software components, and therefore has stronger management capabilities and can meet the needs of modern computer systems.
[0045] In a possible implementation manner, the target DPU card is connected via a PCIE bus, and the DPU card is used to manage at least one of the target DPU cards.
[0046] In a possible implementation, the onboard controller is also used to provide log management, user management, BIOS management, or alarm management.
[0047] In the embodiment of the present application, BMC is the core management component on the DPU mainboard, responsible for managing the interface between the system management software and the platform management hardware, and providing functions such as autonomous monitoring, event recording, and recovery control. BMC can detect the health status of the temperature, voltage, etc. of each component of the DPU card (such as CPU, memory, hard disk, fan, etc.) in real time to ensure the stable operation of the DPU card. If a component of the DPU card is abnormal, BMC can capture the abnormality in the first time, and report the information to the upper-level network management in a timely manner through various industry-wide specifications such as SNMP protocol, SMTP protocol, Redfish protocol, etc., to help operation and maintenance personnel quickly locate the problem. Remote control and management BMC supports remote power on and off and restart of the server, which is convenient for operation and maintenance personnel to operate without contacting the physical server. Through BMC, operation and maintenance personnel can remotely update the firmware and install the system of the server or DPU card to improve work efficiency.
[0048] BMC is a small operating system independent of the server system. It does not rely on other hardware and software on the server, ensuring the separation of business nodes and management nodes and avoiding interference of business nodes on management nodes. BMC uses an independent power supply and remains in working state when the system is shut down, ensuring remote management and control when a server failure occurs.
[0049] In addition to the above general protocols, some server vendors or DPU vendors may customize communication protocols for their specific CPU and BMC designs. These protocols may be extended or modified based on existing standard protocols to meet specific performance, security or management requirements. In summary, the communication protocol between the CPU and BMC chip may vary depending on the server design, vendor implementation and specific requirements. Common protocols include IPMI, middle-layer software protocols (such as ME), I2C and PECI. In actual applications, the communication protocol to be used needs to be determined based on the specific server design and documentation.
[0050] In summary, in the system shown in the embodiment of the present application, the onboard controller is highly integrated with other components of the DPU to form a compact system-on-chip (SoC), which improves system performance and reduces power consumption. Customized requirements can be provided according to scenarios and needs. Remote management functions can also be provided, allowing users to remotely monitor and manage servers through the network, improving the convenience and flexibility of management.
[0051] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, the electronic device 300 includes: a memory 310 and a processor 320 .
[0052] The memory 310 stores a computer program that can be executed by at least one processor 320. The computer program is executed by at least one processor 320 to enable the electronic device to manage the system provided in any of the above embodiments.
[0053] The memory 310 and the processor 320 may be connected via a bus 330 .
[0054] The relevant instructions can be understood by referring to the relevant descriptions and effects corresponding to the method embodiments, which will not be repeated here.
[0055] In the several embodiments provided in the present application, it should be understood that the disclosed system can be implemented in other ways. For example, the system embodiments described above are only schematic, for example, the division of modules is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.
[0056] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the disclosure disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope of the present application is indicated by the claims.
[0057] It should be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.
Claims
1. A DPU card and onboard controller communication management system, characterized in that: include: An onboard controller is integrated on the DPU card, and the onboard controller is used to manage and control the configuration, collection and monitoring of the CPU processor and external sensors on the mainboard; The DPU card and the onboard controller communicate via an I2C or SPI bus.
2. The system according to claim 1, characterized in that The system further includes: the DPU card and the onboard controller are managed and controlled via at least one of a PECI interface, an IPMI interface and a Redfish interface.
3. The system according to claim 2, characterized in that The onboard controller obtains the temperature information and power management information of the CPU through the PECI interface or the IPMI interface.
4. The system according to claim 2, characterized in that The onboard controller remotely monitors and manages the hardware status of the server through the IPMI interface.
5. The system according to claim 2, characterized in that The onboard controller performs remote management of an operation interface through the Redfish interface, and the operation interface is used to query, configure, change or upgrade the DPU card.
6. The system according to any one of claims 1 to 5, characterized in that: The system further comprises: connecting with a target DPU card via a PCIE bus, and the DPU card is used to manage at least one of the target DPU cards.
7. The system according to any one of claims 1 to 5, characterized in that: The onboard controller is also used to provide log management, user management, BIOS management, or alarm management.
8. An electronic device, characterized in that: include: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the electronic device to manage the system as claimed in any one of claims 1 to 7.