Host partition management method and architecture

Abstract

The present application discloses a host partition management method and architecture, which relates to the field of computer technology. The management unit is also divided into multiple management partitions, corresponding to the host partitions of the server, and the host and management units are connected through a logical processing unit. The logical processing unit accurately locates the target device based on the command identifier and the host partition identifier in the first management command of the management partition, and generates a second management command executable by the target device, so that the target device can accurately execute the second management command based on the target device memory address in the second management command. In this way, interface expansion can be achieved through the logical processing unit to meet the target device requirements of multiple host partitions. In addition, the present application only requires one management unit, which can reduce hardware complexity, cost and resource waste. Moreover, the management partition does not need to perform complex logical processing, which can reduce the design complexity of the management partition, shorten its production cycle and reduce its cost.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a host partition management method and architecture. Background Art

[0002] Servers typically need to process large amounts of data. To improve their processing capabilities, servers can be partitioned in terms of hardware. Each partition is equipped with complete computing, storage, networking, and cooling components. These components need to be managed by a baseboard management controller (BMC).

[0003] In related technologies, there are two system architectures for implementing BMC management of partition components. The first solution deploys a BMC for each partition, which increases hardware complexity and cost, and wastes BMC resources. The second solution deploys a BMC for one or more partitions, or even shares a single BMC for all partitions. This results in increased BMC design complexity, a longer production cycle, and higher costs. Summary of the Invention

[0004] This application provides a host partition management method and architecture to at least solve the problems of complexity, high cost, waste of BMC resources, and long production cycle in related technologies.

[0005] The present application provides a host partition management method, which is applied to a host partition management architecture. The host partition management architecture includes: a host, a logical processing unit, and a management unit; the host includes multiple host partitions, and the management unit includes multiple management partitions; the method includes:

[0006] The logic processing unit receives a first management command sent by the management partition, and parses a command identifier and a host partition identifier of the first management command, where the host partition identifier corresponds to the host partition of the management partition;

[0007] The logic processing unit determines a target device memory address corresponding to the first management command according to the command identifier and the host partition identifier;

[0008] The logic processing unit generates a second management command according to the target device memory address, and sends the second management command to the target device of the host partition for execution.

[0009] The present application also provides a host partition management architecture, which includes: a host, a logical processing unit and a management unit; the host includes multiple host partitions, and the management unit includes multiple management partitions; the logical processing unit is used to execute the aforementioned host partition management method.

[0010] Through the present application, the management unit is also divided into multiple management partitions, corresponding to the host partitions of the server, and the host and the management unit are connected through a logical processing unit. The logical processing unit accurately locates the target device based on the command identifier and the host partition identifier in the first management command of the management partition, and generates a second management command executable by the target device, so that the target device can accurately execute the second management command according to the target device memory address in the second management command. In this way, interface expansion can be achieved through the logical processing unit to provide sufficient interfaces to the target devices of the multiple host partitions of the host, meeting the target device requirements of multiple host partitions. In addition, the present application only requires one management unit to implement host partition management, thereby reducing hardware complexity, cost and resource waste. In addition, the logical processing unit of the present application can implement logical processing, and the management partition does not need to perform complex logical processing, thereby reducing the design complexity of the management partition, shortening its production cycle and reducing its cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0012] Figure 1 This is a schematic diagram of a host partition management architecture provided by an embodiment of the present application;

[0013] Figure 2 A flowchart of a host partition management method provided in an embodiment of the present application;

[0014] Figure 3 This is a schematic diagram of the format of a first management command provided in an embodiment of the present application;

[0015] Figure 4 2 is a schematic diagram of the format of a write command in which the second management command provided in an embodiment of the present application is a write command;

[0016] Figure 5 This is a schematic diagram of the format of a second management command provided by an embodiment of the present application, which is a read command;

[0017] Figure 6 This is a schematic diagram of the format of the response message provided in the embodiment of the present application;

[0018] Figure 7 This is a schematic diagram of the format of a channel application command provided in an embodiment of the present application;

[0019] Figure 8This is a data flow diagram in a target device management process provided by an embodiment of the present application. DETAILED DESCRIPTION

[0020] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0021] It should be noted that, in the description of this application, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. The terms "first," "second," etc., in this application are used to distinguish similar objects, and are not used to describe a particular order or sequence.

[0022] In recent years, with the rapid development of the semiconductor industry and integrated circuit technology, servers have become widely used in various data processing scenarios as core equipment in fields such as cloud computing, artificial intelligence, and big data. To improve their processing capabilities, servers can be partitioned in hardware. Each partition can independently run a host system to independently carry out business operations and provide services. Each partition requires complete support for target devices such as computing, storage, network drivers, and heat sinks. The management of these target devices relies on the BMC. As the number of partitions increases, the number of target devices in a server will increase significantly compared to traditional non-partitioned server systems. This poses greater challenges to the server's BMC. Current general-purpose BMCs have limited controller resources and lack sufficient interfaces to manage target devices in multiple host partitions, making it difficult to meet the growing demand for device management.

[0023] To achieve larger-scale target device management in multiple host partition scenarios, one possible solution is to equip each partition with a separate BMC. However, this solution results in each BMC serving only one partition, resulting in redundant and wasted BMC resources. Furthermore, a server requires more BMCs, increasing server costs and hardware complexity. Another possible solution is to use a higher-specification BMC, which typically integrates more interface resources, allowing a single BMC to provide management for multiple partitions simultaneously. However, this solution still has some problems: First, upgrading the BMC's interface resources requires restructuring its overall architecture, which not only significantly increases design complexity but also leads to a significant increase in BMC costs. Second, the design, tape-out, manufacturing, and debugging of the BMC require a significant amount of time, resulting in a long production cycle and high costs, which is not conducive to rapid implementation and widespread application.

[0024] In order to solve the above technical problems, the present application also divides the BMC into multiple partitions, called management partitions, which correspond to the host partitions of the server, and are connected between the host and the BMC through a logical processing unit. The logical processing unit is responsible for data interaction between the host and the BMC. In this way, interface expansion can be achieved through the logical processing unit to provide sufficient interfaces to the target devices of the multiple host partitions of the host to meet the target device requirements of multiple host partitions. In addition, the present application only requires one BMC to achieve management, thereby reducing hardware complexity, cost and resource waste. In addition, the management partition of the present application interacts with the host partition through the routing component, and the BMC does not need to perform complex logical processing. Complex logical processing can be performed by the routing component, thereby reducing the design complexity of the BMC, shortening its production cycle and reducing its cost. On this basis, the logical processing unit can parse the management command to obtain the command identifier and the host partition identifier to determine the target device register address, so that the target device can be accurately managed according to the target device register address.

[0025] Figure 1 This is a schematic diagram of a host partition management architecture provided by an embodiment of the present application. Figure 1As shown, the host partition management architecture includes a host, a logical processing unit, and a management unit. The management unit here can be the aforementioned BMC, and the logical processing unit is connected to the host and BMC respectively. The host includes multiple host partitions, the logical processing unit includes a first transmission component, a routing component, and a second transmission component connected in sequence, and the BMC includes multiple management partitions. The logical processing unit can be implemented by any integrated circuit with logical processing capabilities. For example, commonly used circuits can be CPLDs (Complex Programmable Logic Devices), FPGAs (Field Programmable Gate Arrays), and ASICs (Application Specific Integrated Circuits).

[0026] Each host partition is provided with a target device, which is connected to the second transmission component, thereby realizing the connection between the host partition and the logical processing unit. The management partition and the first transmission component realize the connection between the BMC and the logical processing unit.

[0027] Further, refer to Figure 1 As shown, the second transmission component may include one or more second service processing interfaces for transmitting data between the routing component and the target device. Each second service processing interface may be connected to one or more target devices, and all second service processing interfaces of the logic processing unit are connected to the routing component.

[0028] A management partition includes one or more service management interfaces and a first channel management interface. Accordingly, the first transmission component may include one or more first service processing interfaces and a second channel management interface. The first service processing interface is used to transmit management commands or response data for target devices between the routing component and the service management interface. A first service processing interface is connected to a service management interface. The second channel management interface and the first channel management interface are used to manage the connection channel between the service management interface and the first service processing interface. The second channel management interface is connected to the first channel management interfaces of all management partitions.

[0029] Figure 1 In this example, two host partitions HA1 and HA2, two management partitions MA1 and MA2 are provided. Each host partition has one target device, two second service processing interfaces, and two first service processing interfaces. The number of management partitions is the same as the number of host partitions. Management partition MA1 is used to manage the target device in host partition HA1, and management partition MA2 is used to manage the target device in host partition MA2.

[0030] It is understood that in actual applications, the number of host partitions, management partitions, first service processing interfaces, second service processing interfaces, and service management interfaces can be flexibly configured based on actual scenario requirements. The target device can be any device required for host operation, such as a network card, hard drive, GPU (Graph Processing Unit), temperature sensor, etc.

[0031] based on Figure 1 In the structure shown, when the management partition manages the target device, the management command can reach the target device via the business management interface of the management partition, the first business processing interface, the routing component, and the second business processing interface, and the response data can reach the business management interface of the management partition via the target device, the second business processing interface, the routing component, and the first business processing interface.

[0032] When expanding interfaces through the logical processing circuit, more second service processing interfaces are required. The total number of second service processing interfaces in the logical processing unit may be greater than the total number of service management interfaces in each management partition of the management unit. Thus, even if the management unit provides limited service management interfaces, the additional second service processing interfaces can still connect to target devices in more host partitions, meeting the target device management needs of multiple host partitions.

[0033] Of course, it should be noted that the above-mentioned routing component does not simply transmit data, but needs to parse the data. For its specific role, please refer to the corresponding description of the host management partition management method.

[0034] Finally, it should be noted that the above-mentioned first channel management interface, business management interface, first business processing interface, second channel management interface, and second business processing interface in the embodiment of the present application can be I2C (Inter-Integrated Circuit, integrated circuit bus) interfaces, so the target device can be called an I2C device.

[0035] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0036] based on Figure 1 The host partition management architecture shown in the figure is Figure 2 A flow chart of the host partition management method provided in the embodiment of the present application is shown as follows: Figure 2 As shown, an embodiment of the present application provides a host partition management method, which is described in detail as follows:

[0037] S201: The logic processing unit receives a first management command sent by the management partition, and parses the command identifier and the host partition identifier of the first management command, where the host partition identifier corresponds to the host partition of the management partition.

[0038] When the logical processing unit includes Figure 1 When the first transmission component, routing component and second transmission component are shown, the above-mentioned logical processing unit receives the first management command sent by the management partition, and parses the command identifier and host partition identifier of the first management command, including: the first transmission component receives the first management command sent by the management partition, the first transmission component sends the first management command to the routing component, and the routing component parses the command identifier and host partition identifier of the first management command.

[0039] In some embodiments, the first transmission component includes one or more first business processing interfaces, each of which is connected to a routing component. The first business processing interface is used to transmit the first management command and response data to the first management command. Therefore, the first transmission component receives the first management command sent by the management partition, sends the first management command to the routing component, and the routing component parses the command identifier and host partition identifier of the first management command, including:

[0040] The first service processing interface receives the first management command and parses the slave address of the first management command. If the slave address of the first management command corresponds to the first service processing interface, the first service processing interface sends the first management command to the routing component of the logic processing unit. The routing component parses the command identifier and host partition identifier of the first management command. This prevents the first service processing interface from processing an erroneous command, thereby improving the accuracy of processing the first management command.

[0041] The routing component can extract the slave address from the first management command. When the slave address of the first management command corresponds to the first business processing interface, it means that the first management command is a command sent to the first business processing interface, and the first business processing interface will send the first management command to the routing component for further processing; when the slave address of the first management command does not correspond to the first business processing interface, it means that the first management command is not a command sent to the first business processing interface, and the first business processing interface will not forward the first management command to the routing component for processing.

[0042] The first management command is a command used by the management partition to manage the target device, including but not limited to: write command, read command, modify command, etc. The first management command is generated and sent by a management partition of the management unit, and is used to manage the target device in the host partition corresponding to the management partition. For example, when Figure 1When the management partition MA1 in the host partition HA1 needs to manage the target device in its corresponding host partition HA1, the command identifier and the host partition identifier are determined according to the management requirement, and a first management command is generated according to the command representation and the host partition identifier.

[0043] The command identifier is used to indicate the specific management operation of the first management command. It is a unique identifier for a management operation and may include the target device identifier for the management operation and the management operation identifier for the management operation. In this way, the first management command can be accurately represented based on the command identifier, thereby accurately managing the target device. For example, it can be a concatenation of the target device identifier and the management operation identifier, or some other mapping relationship. The target device identifier indicates the target device to be managed by the first management command, and the management operation identifier indicates the specific management operation content of the first management command.

[0044] Table 1 is a mapping relationship table of a command identifier, a target device identifier, a management operation identifier, and a target device register address provided in an embodiment of the present application. The command identifier in Table 1 is the concatenation result of the target device identifier and the management operation identifier.

[0045]

[0046] Table 1

[0047] As can be seen from Table 1, target device identifier 100 corresponds to a hard drive, target device identifier 101 corresponds to a network card, target device identifier 110 corresponds to a GPU accelerator card, and target device identifier 111 corresponds to a temperature sensor. Different target device identifiers correspond to different target devices. For hard drives, network cards, and temperature sensors, there's no need to separate multiple read contents. Therefore, management operation identifier 00 corresponds to a read operation. For GPU accelerator cards, which have multiple readable contents, management operation identifiers 00 and 01 correspond to reading the accelerator card identifier and manufacturer identifier, respectively.

[0048] The host partition identifier is used to indicate the host partition managed by the first management command and is a unique identifier for the host partition. The host partition identifier and the management partition identifier have a specific mapping relationship, which is used to indicate the relationship between the host partition and the management partition. In some simplest implementations, the host partition and the management partition can establish a corresponding relationship through the identifier. In other words, the management partition is used to manage the host partition with the same identifier.

[0049] Figure 3 This is a schematic diagram of the format of a first management command provided in an embodiment of the present application. Figure 3As shown, the first management command includes, in sequence, a start signal S, slave addresses A6 to A0, a read / write flag 0, target command identifiers B7 to B5, host partition identifiers B4 to B2, management operation identifiers B1 and B0, command parameters D6 to D0, and an end signal P. When the management partition sends the first management command to the logical processing unit, the logical processing unit responds with a response signal to the management partition to confirm successful receipt of the first management command. The response signal includes an acknowledgement signal A.

[0050] Specific reference Figure 3 As shown, after the management partition sends a start signal S, seven-bit slave addresses A6 to A0, and a read / write flag of 0 to the logical processing unit, the logical processing unit replies with an acknowledgment signal A to the management partition, indicating receipt of the start signal S, seven-bit slave addresses A6 to A0, and read / write flag of 0. The read / write flag of 0 sent by the management partition indicates that the management partition will subsequently write target command identifiers B7 to B5, host partition identifiers B4 to B2, and management operation identifiers B1 and B0 to the main line connecting the management partition and the logical processing unit, i.e., the connection channel corresponding to the service management interface that sends the first management command.

[0051] The management partition continues to send target command identifiers B7 to B5, host partition identifiers B4 to B2, and management operation identifiers B1 and B0 to the logical processing unit, and after receiving the confirmation signal A replied by the logical processing unit, continues to send command parameters D6 to D0 to send the data to be written by the first management command to the logical processing unit through the command parameters D6 to D0. When the first management command is a read command, Figure 3 The command parameters in can be invalid data or not transmitted.

[0052] When the management partition receives the confirmation signal A replied by the logic processing unit in response to the command parameters D6 to D0, it sends an end signal P to the logic processing unit, indicating that the sending of the first management command is completed.

[0053] S202: The logic processing unit determines a target device memory address corresponding to the first management command according to the command identifier and the host partition identifier.

[0054] When the logical processing unit includes Figure 1In the case of the first transmission component, routing component, and second transmission component shown, the logic processing unit determines the target device memory address corresponding to the first management command based on the command identifier and the host partition identifier, including: the routing component determines the target device memory address corresponding to the first management command based on the command identifier and the host partition identifier. The target device memory address is an address for storing parameters related to the target device and is also the address to be accessed by the first management command address, including: reading the target device memory address from the target device memory address and / or writing the target device memory address to the target device memory address.

[0055] The target device memory address is a memory address in the target device managed by the first management command, and the first management command is used to write data to and / or read data from the memory address to achieve management of the target device.

[0056] In some embodiments, the routing component determines the target device memory address corresponding to the first management command based on the command identifier and the host partition identifier, including: the routing component searches for the target device memory address from a preset mapping relationship based on the command identifier and the host partition identifier. In this way, the routing component can quickly and accurately determine the target device memory address based on the locally stored preset mapping relationship, thereby ensuring accurate and rapid generation of the second management command, ultimately improving the accuracy and speed of target device management.

[0057] The preset mapping relationship is used to indicate the mapping relationship between the command identifier, the host partition identifier, and the target device memory address. The preset mapping relationship can be a mapping table storing multiple mapping records, each mapping record indicating the mapping relationship between a command identifier, a host partition identifier, and a target device memory address.

[0058] It can be seen from the above preset mapping relationship that the parameter information accessed by different commands in different host partitions is different. For the same device or different devices in different host partitions, due to the different host partition identifiers, even if the command identifiers are the same, the corresponding target device memory addresses are different. For different target devices in the same host partition, due to the different target device identifiers, the command identifiers are different, and thus the corresponding target device memory addresses are also different. For example, referring to Table 1, the hard disk temperature, network card temperature and temperature sensor temperature are read, and the corresponding target device register addresses are all different, namely ADD1, ADD2 and ADD5 respectively. For the same target device in the same host partition, even if the host partition identifier and target device identifier are the same, but the management operation identifiers are different, the command identifiers are different, and thus the corresponding target device memory addresses are also different. For example, referring to Table 1, the GPU accelerator card identifier and the GPU accelerator card manufacturer identifier are read, and the corresponding target device memory addresses are also different, namely ADD3 and ADD4 respectively.

[0059] S203: The logic processing unit generates a second management command according to the memory address of the target device, and sends the second management command to the target device of the host partition for execution.

[0060] The second management command is a command sent by the logic processing unit to the target device and is generated according to the first management command. The second management command carries at least the memory address of the target device.

[0061] In one possible implementation, the above-mentioned logical processing unit generates a second management command based on the target device memory address and sends it to the target device of the host partition for execution, including: the routing component determines the target format of the second management command based on the command identifier; the routing component parses the first management command to obtain the command parameters; the routing component generates the second management command in the target format based on the target device memory address and the command parameters, and then the routing component sends the second management command in the target format to the target device of the host partition for execution. In this way, the embodiment of the present application can further limit the management operation on the target device in combination with the command parameters, which can improve the accuracy of the management operation. In addition, the present application can also identify the type of the second management command based on the command identifier, and then generate a command in the corresponding format, which can realize the execution of multiple types of commands and diversify the management of the target device.

[0062] It is understood that different commands may correspond to different target formats. The target formats here primarily include read command formats and write command formats. In other words, the formats of the second management command as a read command and the second management command as a write command are different. For example, the lengths and / or contents of the two commands may differ. A write command may be longer than a read command, and a write command may also include command parameters to be written.

[0063] In one possible implementation, determining the target format of the second management command based on the command identifier includes: parsing the command identifier to obtain a management operation identifier, and determining whether the first management command is a write command or a read command based on the management operation identifier, wherein the second management command and the first management command are of the same type. For specific correspondence, see Table 1. After determining whether the second management command is a read command or a write command, a target format of the second management type may be determined based on the type of the second management command. The mapping relationship between the target format and the type of the second management command may be preset.

[0064] Figure 4 1 is a schematic diagram of the format of the second management command provided in the embodiment of the present application as a write command. Figure 4 As shown, the second management command includes, in sequence, a start signal S, a seven-bit slave address A6 to A0, a read / write flag 0, target device memory addresses B7 to B0, command parameters D6 to D0, and an end signal P. When the logical processing unit sends the second management command to the host partition, the host partition responds with a response signal to the logical processing unit in response to the second management command to confirm successful receipt of the second management command. The response signal includes an acknowledgement signal A.

[0065] Specific reference Figure 4 As shown, after the logical processing unit sends a start signal S, a seven-bit slave address A6 to A0, and a read / write flag of 0 to the target device of the host partition, the target device replies with an acknowledgment signal to the logical processing unit, indicating receipt of the start signal S, the seven-bit slave address A6 to A0, and the read / write flag of 0. The read / write flag of 0 sent by the logical processing unit indicates that the logical processing unit will subsequently write to the target device memory addresses B7 to B0 on ​​the main line connecting the target device and the logical processing unit, that is, the connection channel corresponding to the second service processing interface that sends the second management command.

[0066] The logic processing unit continues to send the target device register address B7 to B0 to the target device, and after receiving the confirmation signal A replied by the target device, continues to send command parameters D6 to D0 to send the data to be written by the second management command to the target device through command parameters D6 to D0.

[0067] When the logic processing unit receives the confirmation signal A replied by the target device in response to the command parameters D6 to D0, it sends an end signal P to the target device, indicating that the sending of the second management command is completed.

[0068] Figure 5 This is a schematic diagram of the format of a second management command provided by an embodiment of the present application, which is a read command. Figure 5As shown, the second management command sequentially includes a start signal S, slave addresses A6 to A0, a read / write flag 0, target device memory addresses B7 to B0, slave addresses A6 to A0, a read / write flag 1, a confirmation signal, and an end signal P. When the logical processing unit sends the second management command to the host partition, the host partition responds to the logical processing unit with a response signal in response to the second management command. The response signal includes an confirmation signal A and response data D7 to D0.

[0069] After the logical processing unit sends a start signal S, slave addresses A6 to A0, and a read / write flag of 0 to the target device in the host partition, the target device replies with an acknowledgment signal A to the logical processing unit, indicating receipt of the start signal S, the seven-bit slave addresses A6 to A0, and the read / write flag of 0. The read / write flag of 0 sent by the logical processing unit indicates that the logical processing unit will subsequently write to the target device memory addresses B7 to B0 on ​​the main line connecting the target device and the logical processing unit, i.e., the connection channel corresponding to the second service processing interface for sending the second management command.

[0070] The logic processing unit continues to send the target device register addresses B7 to B0 to the target device, and after receiving the confirmation signal A in response from the target device, continues to send the slave addresses A6 to A0 and the read / write flag 1. The read / write flag 1 is used to instruct the logic processing unit to subsequently read the response data from the main line, which is the main line connecting the target device and the logic processing unit, that is, the connection channel corresponding to the second service processing interface for sending the second management command.

[0071] After the target device responds with the confirmation signal A to the addresses A6 to A0, it sends the response data D7 to D0 to the logic processing unit.

[0072] After sending a confirmation signal A for the response data to the target device, the logic processing unit sends an end signal P, indicating that the sending of the second management command is completed.

[0073] It is understandable that when the second management command is determined to be a write command, Figure 4 When the second management command is determined to be a read command, the second management command can be generated according to the target format shown. Figure 5 The target format shown generates a second management command. Figure 4 or Figure 5 When the second management command is shown, it can be determined whether the command is a read command or a write command according to the format, and then the second management command is correctly executed.

[0074] The command parameter is the data content written to the memory address of the target device. Therefore, when the second management command is a read command, the second management command does not include the command parameter. When the second management command is a write command, the second management command includes the command parameter. The command parameter is parsed from the first management command.

[0075] In some possible implementations, the routing component sends the second management command to the target device of the host partition for execution, including: the routing component parses the command identifier to obtain the target device identifier, and sends the second management command to the target device of the host partition for execution through the host partition identifier and the second business processing interface corresponding to the target device identifier. The second business processing interface is the interface to which the target device managed by the first management command is connected, and different target devices in different host partitions are connected to different second business processing interfaces. In the embodiment of the present application, the second business processing interface is determined based on the host partition identifier and the target device identifier to ensure that the second management command is sent to the target device of the corresponding host partition, thereby avoiding errors in sending the second management command and improving the management accuracy of the target device.

[0076] Reference Figure 3 As shown, the routing component can extract the target device identifier from the first management command according to the format of the first management command, and extract the second business processing interface from the device interface mapping relationship stored locally in the routing component. The device interface mapping relationship here includes the mapping relationship between the host partition identifier, the target device identifier and the second business processing interface.

[0077] In some embodiments, the routing component generates a second management command in a target format based on the target device memory address and command parameters, including: the routing component determines the slave address of the second management command based on the target device identifier; the routing component generates the second management command in the target format based on the slave address of the second management command, the target device storage address, and the command parameters. Accordingly, when the target device of the host partition receives the second management command, it parses the slave address of the second management command; when the slave address of the second management command corresponds to the target device, it executes the second management command; when the slave address of the second management command does not correspond to the target device, it does not execute the second management command. The embodiment of the present application can verify the second management command based on the slave address of the second management command, avoid errors in sending the second management command, and help improve management accuracy.

[0078] The slave address of the second management command is used to indicate the target device receiving the second management command, which can be the address of the target device or the target device identifier. The relationship between the target device identifier and the address of the target device can be pre-stored in the routing component.

[0079] After determining the slave address, target device storage address and command parameters of the second management command, the routing component Figure 4 or Figure 5 The target format shown generates a corresponding second management command.

[0080] It is understood that when the first management command and the second management command are write commands, the target device does not need to return response data to the logic processing device after executing the second management command. However, when the first management command and the second management command are read commands, the target device needs to return response data to the logic processing device after executing the second management command. The response data can be generated and transmitted through the following steps:

[0081] First, after the routing component generates the second management command based on the first management command, it records the command mapping relationship between the first and second management commands for subsequent generation of a response message. Then, the second service processing interface receives the response data sent by the target device and forwards it to the routing component. Finally, the routing component generates a response message based on the command mapping relationship and the response data, and sends the response message to the management partition via the first service processing interface of the logical processing unit. In this way, the response message can include the same content as the first management command, so that after receiving the response message, the management partition can establish the relationship between the first management command and the response message for subsequent processing of the response message, thereby improving the accuracy of subsequent processing of the response data.

[0082] Among them, the response data here is Figure 5 The response data shown, the routing component can be based on Figure 6 The format shown generates a response message, that is, the routing component needs to determine the first management command corresponding to the second management command from the command mapping relationship, and then the response data of the second management command and the command identifier and host partition identifier in the first management command are mapped according to the command mapping relationship. Figure 6 The format shown is added to the response message.

[0083] In some embodiments, the routing component generates a response message based on the command mapping relationship and the response data, including: the routing component determines the corresponding first management command from the command mapping relationship based on the second management command corresponding to the response data; then, the routing component extracts the command identifier and host partition identifier from the first management command, and generates a response message based on the command identifier, host partition identifier, and response data. After receiving the response message, the management partition can establish a mapping relationship between the response message and the first management command based on the command identifier and host partition identifier therein, so that the management partition can subsequently process the response data in the response message in the software interface that calls the first management command. In this way, the accurate processing of the response data can be ensured, improving the accuracy of management of the target device.

[0084] It is understood that the routing component and the target device are connected in accordance with Figure 5 After the routing component sends the start signal, slave address, read / write flag, target device memory address, slave address and read / write flag to the target device in sequence, the routing component waits for the response data for the second management command.

[0085] In some possible implementations, generating a response message based on the command identifier, the host partition identifier, and the response data includes: the routing component determining a slave address of the response message based on the first channel identifier; and the routing component generating the response message based on the slave address of the response message, the command identifier, the host partition identifier, and the response data. In this way, the slave address of the response data can be used to indicate the management partition, thereby ensuring accurate transmission and processing of the response data.

[0086] It is understood that the routing component can be Figure 6 The format shown in FIG. 1 adds the slave address, command identifier, host partition identifier, and response data of the response message to the response data.

[0087] Figure 6 This is a schematic diagram of the format of the response message provided in the embodiment of this application. Figure 6 As shown, the response message includes, in sequence, a start signal S, slave addresses A6 to A0, a read / write flag 0, target command identifiers B7 to B5, host partition identifiers B4 to B2, management operation identifiers B1 and B0, response data D7 to D0, an acknowledgment signal A, and an end signal P. During the process of the logical processing unit sending the second management command to the management partition, the management partition will reply to the logical processing unit with a response signal in response to the second response message to confirm that the response message has been successfully received. The response signal includes an acknowledgment signal A.

[0088] After the logical processing unit sends a start signal S, slave addresses A6 to A0, and a read / write flag of 0 to the management partition, the management partition replies with an acknowledgment signal A to the logical processing unit, indicating receipt of the start signal S, the seven-bit slave addresses A6 to A0, and the read / write flag of 0. The read / write flag of 0 sent by the logical processing unit indicates that the logical processing unit will subsequently write the target command identifiers B7 to B5, host partition identifiers B4 to B2, and management operation identifiers B1 and B0 to the main line connecting the logical processing unit and the management partition, i.e., the connection channel corresponding to the first service processing interface that sends the response message.

[0089] The logical processing unit continues to send target command identifiers B7 to B5, host partition identifiers B4 to B2, and management operation identifiers B1 and B0 to the management partition, and after receiving the confirmation signal A replied by the management partition, continues to send response data D7 to D0.

[0090] After the management partition replies with the confirmation signal A for the response data D7 to D0, the logic processing unit sends an end signal, indicating that the response message is sent successfully.

[0091] Before the routing component sends the response message to the management partition, the routing component also needs to apply for a channel from the management partition. Specifically, the routing component generates a response message based on the command mapping relationship and the response data, and sends the response message to the management partition through the first business processing interface of the logical processing unit, including:

[0092] The second channel management interface of the logic processing unit sends a second channel request command to the first channel management interface of the management partition. The second channel management interface receives the first channel identifier sent by the first channel management interface and forwards it to the routing component. The routing component generates a response message based on the command mapping relationship and the response data, and sends the response message to the management partition via the first service processing interface corresponding to the first channel identifier. This allows the response message to be transmitted over an idle channel as much as possible, which helps improve the success rate of response message transmission.

[0093] The second channel request command is used to request a transmission channel for the response data. The transmission channel for the response data is a connection channel between the first service processing interface in the logical processing unit and the service management interface in the management partition. There are multiple first service processing interfaces in the logical processing unit, and multiple service management interfaces in the management partition. Each first service processing interface is connected to a service management interface to form a connection channel. Therefore, there are multiple connection channels between the logical processing unit and the management partition, and the first channel identifier is used to indicate one of the connection channels.

[0094] After the first channel management interface generates the first channel identifier, the corresponding first service processing interface can be determined. The routing component has a mapping relationship between the first channel identifier, the first service processing interface, and the service management interface.

[0095] In some embodiments, the second channel management interface of the logical processing unit sends a second channel request command to the first channel management interface of the management partition, including: the second channel management interface of the logical processing unit determines the first channel management interface identifier of the corresponding management partition based on the host partition identifier, using the identifier as the slave address of the second channel request command; the second channel management interface generates the second channel request command based on the slave address and the channel request command identifier of the second channel request command; and the second channel management interface sends the second channel request command to the first channel management interface of the management partition. In this way, the first channel management interface can identify the second channel request command based on the slave address and the channel request command identifier in the second channel request command, and then accurately execute the second channel request command, thereby helping to ensure channel allocation accuracy.

[0096] Specifically, the mapping relationship between the host partition identifier and the first channel management interface identifier of the management partition is predetermined and stored in the second channel management interface.

[0097] In some possible implementations, before the second channel management interface receives the first channel identifier sent by the first channel management interface, the first channel management interface can generate the first channel identifier through the following process: the first channel management interface parses the slave address in the second channel request command; when the slave address of the second channel request command corresponds to the first channel management interface, the first channel management interface parses the command identifier of the second channel request command; when the command identifier of the second channel request command is a channel request command identifier, the first channel management interface generates the first channel identifier and sends it to the second channel management interface. In this way, when the first channel management interface determines that the second channel request command is correct, it assigns the first channel identifier, which can ensure the accuracy of channel allocation, avoid applying for channel allocation for abnormal channels, and ensure the full and effective utilization of channels.

[0098] It is understandable that when the slave address of the second channel request command corresponds to the first channel management interface, it represents that the second channel request command is a command sent to the first channel management interface. Furthermore, when the command identifier of the second channel request command is a channel request command identifier, it represents that the second channel request command is a command for requesting a channel. At this time, the first channel management interface generates a first channel identifier. Of course, if the slave address of the second channel request command does not correspond to the first channel management interface, and / or the command identifier of the second channel request command is not a channel request command identifier, the first channel management interface deems the second channel request command to be an abnormal command and does not process it.

[0099] Figure 7 This is a schematic diagram of the format of a channel application command provided in an embodiment of the present application. Figure 7 As shown, when Figure 7 When the channel request shown is a second channel request command, the second channel request command sequentially includes a start signal S, slave addresses A6 to A0, a read / write flag 0, channel request command identifiers B7 to B0, slave addresses A6 to A0, a read / write flag 1, an acknowledgement signal A, and an end signal P. When the second channel management interface sends the second channel request command to the first channel management interface, the first channel management interface sends a response signal to the second channel management interface in response to the second channel request command. The response signal includes an acknowledgement signal A and channel identifiers D7 to D0, corresponding to the second channel request command, where the channel identifier is the first channel identifier.

[0100] Specific reference Figure 7As shown, after the second channel management interface sends a start signal S, a seven-bit slave address A6 to A0, and a read / write flag of 0 to the first channel management interface, the first channel management interface replies with an acknowledgment signal A to the second channel management interface, indicating receipt of the start signal S, the seven-bit slave address A6 to A0, and the read / write flag of 0. The read / write flag of 0 sent by the second channel management interface indicates that the second channel management interface will subsequently write channel request command identifiers B7 to B0 to the main line connecting the second channel management interface and the first channel management interface.

[0101] The second channel management interface continues to send channel request command identifiers B7 to B0 to the first channel management interface, and after receiving the confirmation signal A replied by the first channel management interface, continues to send slave addresses A6 to A0 and read / write flag 1. The read / write flag 1 is used to indicate that the second channel management interface is about to read the channel identifier on the main line.

[0102] After replying the confirmation signal A to the second channel management interface, the first channel management interface sends the first channel identifier to the second channel management interface.

[0103] After receiving the first channel identifier, the second channel management interface will reply a confirmation signal A and an end signal P to the first channel management interface, indicating that the sending of the second channel request command is completed.

[0104] In some embodiments, the first channel management interface generates a first channel identifier by: obtaining a channel status between the management partition and the logical processing unit, the channel status including the channel identifier and the number of pending commands; and then selecting the channel identifier with the smallest number of pending commands as the first channel identifier. This minimizes the latency of sending response messages and improves management efficiency.

[0105] The first channel management interface of a management partition can obtain the channel status of all connection channels between the management partition and the logical processing unit, specifically the channel status between the service management interface in the management partition and the first service processing interface in the logical processing unit. A service management interface is connected to a first service processing interface to form a connection channel. A channel identifier uniquely identifies a connection channel; that is, one channel identifier corresponds to one first service processing interface and one service management interface.

[0106] In scenarios with high traffic volume, the connection channel between the logical processing unit and the management partition may be busy. Each connection channel has a corresponding pending command queue, which contains response messages for that connection channel. To minimize response message transmission latency, the channel ID with the smallest number of pending commands can be selected as the first channel ID.

[0107] It can be understood that when there are multiple channel identifiers with the smallest number of commands to be processed, a channel identifier can be randomly selected from the multiple channel identifiers as the first channel identifier.

[0108] In some possible implementations, upon receiving a response message, the service management interface of the management partition parses the slave address of the response message. If the slave address of the response message corresponds to a service management interface, it indicates that the response message is intended for that service management interface, and the service management interface processes the response message. If the slave address of the response message does not correspond to a service management interface, it indicates that the response message is not intended for that service management interface, and the service management interface does not process the response message. This ensures that the response data is received and processed correctly, thereby ensuring accurate management of the target device.

[0109] The business management interface's processing of the received response data may include: reading the host partition identifier and command identifier, and response data from the response message, and determining the corresponding first management command based on the host partition identifier and command identifier, and finally returning the response data to the software interface that initiated the first management command for subsequent processing of the response data.

[0110] In some possible implementations, before the logical processing unit receives the first management command sent by the management partition, the management partition needs to apply for a channel from the logical processing unit. The management partition is provided with a first channel management interface, and the logical processing unit is provided with a second channel management interface, and the first channel management interface is connected to the second channel management interface. Based on the connection relationship between the first channel management interface and the second channel management interface, the process of the management partition applying for a channel includes the following steps: first, the first channel management interface of the management partition sends a first channel request command to the second channel management interface of the logical processing unit; then, the first channel management interface receives the second channel identifier sent by the second channel management interface; finally, the management partition sends the first management command to the logical processing unit via the service management interface corresponding to the second channel identifier. In this way, it can be ensured that the first management command is transmitted through an idle service management interface, which helps to improve the transmission success rate of the first management command and thereby improve the management success rate of the target device.

[0111] The second channel identifier is the channel identifier assigned by the first channel management interface to the first channel request command. The connection channel corresponding to the second channel identifier is used to transmit the first management command from the service management interface of the management partition to the first service processing interface in the logical processing unit. Here, the service management interface identifier, the first service processing interface identifier, and the second channel identifier have a one-to-one correspondence. Therefore, after determining the second channel identifier, the service management interface for transmitting the first management command can be determined.

[0112] It should be noted that the slave address in the first management command corresponds to the first business processing interface corresponding to the second channel identifier in the logical processing unit. Therefore, when the first business interface receives the first management command, it can determine whether it is the first management command sent to itself based on the slave address, thereby ensuring the successful reception of the first management command.

[0113] In some possible implementations, before the first channel management interface receives the second channel identifier sent by the second channel management interface, the second channel management interface can generate the second channel identifier through the following process: the second channel management interface parses the slave address in the first channel request command; and when the slave address of the first channel request command corresponds to the second channel management interface, the second channel management interface parses the command identifier of the first channel request command; when the command identifier of the first channel request command is a channel request command identifier, the second channel management interface generates the second channel identifier and sends it to the first channel management interface. In this way, when the second channel management interface determines that the first channel request command is correct, it assigns the second channel identifier, which can ensure the accuracy of channel allocation, avoid applying for channel allocation for abnormal channels, and ensure the full and effective utilization of the channel.

[0114] It is understandable that when the slave address of the first channel request command corresponds to the second channel management interface, it represents that the first channel request command is a command sent to the second channel management interface. Furthermore, when the command identifier of the first channel request command is a channel request command identifier, it represents that the first channel request command is a command for requesting a channel. At this time, the second channel management interface generates a second channel identifier. Of course, if the slave address of the first channel request command does not correspond to the second channel management interface, and / or the command identifier of the first channel request command is not a channel request command identifier, the second channel management interface deems the first channel request command to be an abnormal command and does not process it.

[0115] In some possible implementations, the second channel management interface generates the second channel identifier, including: the second channel management interface parsing the management partition identifier of the first channel request command and determining the second channel identifier based on the management partition identifier. In this way, the second channel management interface can allocate the connection channel between the logical processing unit and the management partition that initiated the first channel request command to the management partition that initiated the first channel request command based on the management partition identifier carried in the first channel request command, thereby avoiding channel allocation errors that could lead to transmission failure of the first management command, ensuring successful transmission of the first management command, and thereby improving the success rate of target device management.

[0116] When the first channel management interfaces of multiple management partitions are connected to the same second channel management interface, if the first channel management interface of a management partition needs to send a first channel request command to the logical processing unit, the first channel management interface of the management partition needs to add its own management partition identifier to the first channel request so that the second channel management interface of the logical processing unit can allocate a connection channel between the management partition and the logical processing unit based on the management partition identifier, thereby avoiding channel allocation errors.

[0117] Reference Figure 7 As shown, when Figure 7 When the channel request shown is a first channel request command, the first channel request command includes, in sequence, a start signal S, slave addresses A6 to A0, a read / write flag 0, channel request command identifiers B7 to B0, slave addresses A6 to A0, a read / write flag 1, an acknowledgement signal A, and an end signal P. When the first channel management interface sends the first channel request command to the second channel management interface, the second channel management interface sends a response signal to the first channel management interface in response to the first channel request command. The response signal includes an acknowledgement signal A and channel identifiers D7 to D0, corresponding to the first channel request command, where the channel identifier is the second channel identifier.

[0118] Specific reference Figure 7 As shown, after the first channel management interface sends a start signal S, a seven-bit slave address A6 to A0, and a read / write flag of 0 to the second channel management interface, the second channel management interface replies with an acknowledgment signal A to the first channel management interface, indicating receipt of the start signal S, the seven-bit slave address A6 to A0, and the read / write flag of 0. The read / write flag of 0 sent by the first channel management interface indicates that the first channel management interface will subsequently write channel request command identifiers B7 to B0 to the main line connecting the first channel management interface and the second channel management interface.

[0119] The first channel management interface continues to send channel request command identifiers B7 to B0 to the second channel management interface, and after receiving the confirmation signal A replied by the second channel management interface, continues to send slave addresses A6 to A0 and read / write flag 1. The read / write flag 1 is used to indicate that the first channel management interface will read the second channel identifier from the main line.

[0120] After replying the confirmation signal A to the first channel management interface, the second channel management interface sends the second channel identifier to the first channel management interface.

[0121] After receiving the second channel identifier, the first channel management interface will reply a confirmation signal and an end signal to the second channel management interface, indicating that the sending of the first channel request command is completed.

[0122] In some possible implementations, the second channel management interface determines the second channel identifier based on the management partition identifier, including: first, the second channel management interface obtains the channel status between the management partition corresponding to the management partition identifier and the logical processing unit, where the channel status includes the channel identifier and the number of pending commands; then, the second channel management interface uses the channel identifier with the smallest number of pending commands as the second channel identifier. This can minimize the latency of sending the first management command and improve management efficiency.

[0123] The second channel management interface can obtain the channel status of all connection channels between the logical processing unit and the management partition corresponding to the management partition identifier, specifically, the channel status of the connection channel from the first service processing interface in the logical processing unit to the service management interface. A first service processing interface is connected to a service management interface to form a connection channel. The channel identifier uniquely identifies a connection channel; that is, one channel identifier corresponds to one first service processing interface and one service management interface.

[0124] In scenarios with high traffic volume, the connection channel between the management partition and the logical processing unit may be busy. Each connection channel has a corresponding pending command queue containing pending commands for that connection channel. Therefore, to minimize the delay in sending the first management command, the channel identifier with the smallest number of pending commands can be selected as the second channel identifier.

[0125] It can be understood that when there are multiple channel identifiers with the smallest number of commands to be processed, one channel identifier can be randomly selected from the multiple channel identifiers as the second channel identifier.

[0126] Figure 8 This is a data flow diagram of a target device management process provided by an embodiment of the present application. Figure 8 As shown in the figure, during the target device management process, the data flow process includes the following steps:

[0127] S301: A first channel management interface of a management partition sends a first channel request command to a second channel management interface of a logical processing unit.

[0128] S302: The first channel management interface receives a second channel identifier sent by the second channel management interface.

[0129] S303: The service management interface corresponding to the second channel identifier sends a first management command to the first service processing interface.

[0130] S304: The first service processing interface forwards the first management command to the routing component.

[0131] S305: The routing component sends the second management command mapped by the first management command to the second service processing interface.

[0132] S306: The second service processing interface sends the second management command to the target device of the host partition.

[0133] S307: The target device sends response data of the second management command to the second service processing interface.

[0134] S308: The second service processing interface sends the response data to the routing component.

[0135] S309: The routing component sends a second channel request command to the first channel management interface through the second channel management interface.

[0136] S309: The first channel management interface sends the first channel identifier to the routing component through the second channel management interface.

[0137] S310: The routing component sends a response message corresponding to the response data to the service management interface through the first service processing interface corresponding to the first channel identifier.

[0138] Corresponding to the above method, the present application further provides a host partition management architecture, which includes: a host, a logical processing unit, and a management unit; the host includes multiple host partitions, and the management unit includes multiple management partitions; the logical processing unit is used to:

[0139] Receive a first management command sent by the management partition, and parse the command identifier and host partition identifier of the first management command, where the host partition identifier corresponds to the host partition of the management partition; determine the target device memory address corresponding to the first management command based on the command identifier and the host partition identifier; generate a second management command based on the target device memory address, and send it to the target device of the host partition for execution.

[0140] In some possible implementations, the first service processing interface of the logic processing unit is used to:

[0141] Receive a first management command, parse the slave address of the first management command, and, when the slave address of the first management command corresponds to the first business processing interface, send the first management command to the routing component of the logical processing unit; the routing component is used to parse the command identifier and host partition identifier of the first management command.

[0142] In some possible implementations, the routing component is also used to:

[0143] According to the command identifier and the host partition identifier, the target device memory address is searched from a preset mapping relationship, where the preset mapping relationship is used to indicate a mapping relationship among the command identifier, the host partition identifier, and the target device memory address.

[0144] In some possible implementations, the routing component is also used to:

[0145] Determine the target format of the second management command according to the command identifier; parse the first management command to obtain command parameters; generate the second management command in the target format according to the target device memory address and command parameters, and send it to the target device of the host partition for execution.

[0146] In some possible implementations, the routing component is also used to:

[0147] The command identifier is parsed to obtain the target device identifier, and the second management command is sent to the target device of the host partition for execution through the second service processing interface corresponding to the host partition identifier and the target device identifier.

[0148] In some possible implementations, the routing component is also used to:

[0149] Determine the slave address of the second management command according to the target device identifier; generate the second management command in the target format according to the slave address of the second management command, the target device storage address and the command parameters;

[0150] The target device of the host partition is used to: parse the slave address of the second management command when receiving the second management command; and execute the second management command when the slave address of the second management command corresponds to the target device.

[0151] In some possible implementations, the routing component is also used to:

[0152] Record the command mapping relationship between the first management command and the second management command; the second business processing interface receives the response data sent by the target device and forwards it to the routing component; based on the command mapping relationship and the response data, a response message is generated, and the response message is sent to the management partition through the first business processing interface of the logical processing unit.

[0153] In some possible implementations, the second channel management interface of the logic processing unit is further configured to:

[0154] Sending a second channel request command to the first channel management interface of the management partition; receiving the first channel identifier sent by the first channel management interface and forwarding it to the routing component;

[0155] The routing component generates a response message according to the command mapping relationship and the response data, and sends the response message to the management partition through the first service processing interface corresponding to the first channel identifier.

[0156] In some possible implementations, the routing component is also used to:

[0157] According to the second management command corresponding to the response data, the corresponding first management command is determined from the command mapping relationship; the command identifier and host partition identifier in the first management command are extracted, and a response message is generated according to the command identifier, host partition identifier and response data.

[0158] In some possible implementations, the routing component is also used to:

[0159] The slave address of the response message is determined according to the first channel identifier; and the response message is generated according to the slave address, command identifier, host partition identifier and response data of the response message.

[0160] In some possible implementations, the second channel management interface of the logic processing unit is further configured to:

[0161] Determine the first channel management interface identifier of the corresponding management partition according to the host partition identifier, and use it as the slave address of the second channel request command; generate the second channel request command according to the slave address and the channel request command identifier of the second channel request command; and send the second channel request command to the first channel management interface of the management partition.

[0162] In some possible implementations, the first channel management interface is further configured to:

[0163] Before the second channel management interface receives the first channel identifier sent by the first channel management interface, the second channel management interface parses the slave address in the second channel request command; when the slave address of the second channel request command corresponds to the first channel management interface, the second channel request command is parsed; when the command identifier of the second channel request command is the channel request command identifier, the first channel identifier is generated and sent to the second channel management interface.

[0164] In some possible implementations, the first channel management interface is further configured to:

[0165] The channel status between the management partition and the logical processing unit is obtained, where the channel status includes a channel identifier and the number of pending commands; and the channel identifier with the smallest number of pending commands is used as the first channel identifier.

[0166] In some possible implementations, the service management interface of the management partition is further used to:

[0167] When a response message is received, the slave address of the response message is parsed; when the slave address of the response message corresponds to a service management interface, the response data is processed.

[0168] In some possible implementations, the first channel management interface of the management partition is further used to:

[0169] Before the logical processing unit receives the first management command sent by the management partition, the logical processing unit sends a first channel request command to the second channel management interface of the logical processing unit; and receives a second channel identifier sent by the second channel management interface;

[0170] The management partition is further configured to send the first management command to the logic processing unit via the service management interface corresponding to the second channel identifier, wherein the slave address in the first management command corresponds to the first service processing interface corresponding to the second channel identifier in the logic processing unit.

[0171] In some possible implementations, the second channel management interface is further used to:

[0172] Before the first channel management interface receives the second channel identifier sent by the second channel management interface, the slave address in the first channel request command is parsed; when the slave address of the first channel request command corresponds to the second channel management interface, the command identifier of the first channel request command is parsed; when the command identifier of the first channel request command is the channel request command identifier, a second channel identifier is generated and sent to the first channel management interface.

[0173] In some possible implementations, the second channel management interface is further used to:

[0174] The management partition identifier of the first channel request command is parsed, and the second channel identifier is determined according to the management partition identifier.

[0175] In some possible implementations, the second channel management interface is further used to:

[0176] The channel status between the management partition corresponding to the management partition identifier and the logical processing unit is obtained, where the channel status includes the channel identifier and the number of pending commands; and the channel identifier with the smallest number of pending commands is used as the second channel identifier.

[0177] In some possible implementations, the command identifier includes: a target device identifier and a management operation identifier.

[0178] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method.

[0179] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0180] The above describes in detail the host partition management method and architecture provided by this application. This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only intended to help understand the method and core concept of this application. It should be noted that for those skilled in the art, without departing from the principles of this application, various improvements and modifications can be made to this application, and such improvements and modifications also fall within the scope of protection of the claims of this application.

Claims

1. A host partition management method, characterized in that: The method is applied to a host partition management architecture, the host partition management architecture comprising: a host, a logical processing unit, and a management unit; the host comprises a plurality of host partitions, and the management unit comprises a plurality of management partitions; the method comprises: The logic processing unit receives a first management command sent by the management partition, and parses a command identifier and a host partition identifier of the first management command, where the host partition identifier corresponds to a host partition of the management partition, and the host partition identifier is used to indicate the host partition managed by the first management command; The logic processing unit determines, according to the command identifier and the host partition identifier, a target device memory address corresponding to the first management command; The logic processing unit generates a second management command according to the target device memory address, and sends the second management command to the target device of the host partition for execution.

2. The method according to claim 1, characterized in that The logic processing unit receives a first management command sent by the management partition, and parses a command identifier and a host partition identifier of the first management command, including: The first service processing interface of the logic processing unit receives the first management command and parses the slave address of the first management command; When the slave address of the first management command corresponds to the first service processing interface, the first service processing interface sends the first management command to the routing component of the logical processing unit; The routing component parses the command identifier and the host partition identifier of the first management command.

3. The method according to claim 2, characterized in that The logic processing unit determines, according to the command identifier and the host partition identifier, a target device memory address corresponding to the first management command, including: The routing component searches for the target device memory address from a preset mapping relationship according to the command identifier and the host partition identifier, where the preset mapping relationship is used to indicate the mapping relationship among the command identifier, the host partition identifier, and the target device memory address.

4. The method according to claim 3, characterized in that The logic processing unit generates a second management command according to the target device memory address and sends the second management command to the target device of the host partition for execution, including: The routing component determines a target format of the second management command according to the command identifier; The routing component parses the first management command to obtain command parameters; The routing component generates a second management command in the target format according to the target device memory address and the command parameters, and sends the second management command to the target device of the host partition for execution.

5. The method according to claim 4, characterized in that The target device sending the data to the host partition for execution includes: The routing component parses the command identifier to obtain a target device identifier, and sends the second management command to the target device of the host partition for execution through the host partition identifier and a second service processing interface corresponding to the target device identifier.

6. The method according to claim 5, characterized in that The routing component generates a second management command in the target format according to the target device memory address and the command parameter, including: The routing component determines a slave address of the second management command according to the target device identifier; The routing component generates the second management command in the target format according to the slave address of the second management command, the target device storage address, and the command parameters; The method further comprises: When receiving the second management command, the target device of the host partition parses the slave address of the second management command; When the slave address of the second management command corresponds to the target device, the second management command is executed.

7. The method according to claim 6, characterized in that The method further comprises: The routing component records a command mapping relationship between the first management command and the second management command; The second service processing interface receives the response data sent by the target device and forwards it to the routing component; The routing component generates a response message according to the command mapping relationship and the response data, and sends the response message to the management partition through the first service processing interface of the logical processing unit.

8. The method according to claim 7, characterized in that The routing component generates a response message according to the command mapping relationship and the response data, and sends the response message to the management partition through the first service processing interface of the logical processing unit, including: The second channel management interface of the logical processing unit sends a second channel request command to the first channel management interface of the management partition; The second channel management interface receives the first channel identifier sent by the first channel management interface and forwards it to the routing component; The routing component generates the response message according to the command mapping relationship and the response data, and sends the response message to the management partition through the first service processing interface corresponding to the first channel identifier.

9. The method according to claim 8, characterized in that The routing component generates the response message according to the command mapping relationship and the response data, including: The routing component determines, according to the second management command corresponding to the response data, the corresponding first management command from the command mapping relationship; The routing component extracts the command identifier and the host partition identifier in the first management command, and generates the response message according to the command identifier, the host partition identifier and the response data.

10. The method according to claim 9, characterized in that Generating the response message according to the command identifier, the host partition identifier, and the response data includes: The routing component determines a slave address of the response message according to the first channel identifier; The routing component generates the response message according to the slave address of the response message, the command identifier, the host partition identifier and the response data.

11. The method according to claim 9, characterized in that The second channel management interface of the logical processing unit sends a second channel request command to the first channel management interface of the management partition, including: The second channel management interface of the logic processing unit determines the first channel management interface identifier of the corresponding management partition according to the host partition identifier as the slave address of the second channel request command; The second channel management interface generates the second channel request command according to the slave address and the channel request command identifier of the second channel request command; The second channel management interface sends the second channel request command to the first channel management interface of the management partition.

12. The method according to claim 11, characterized in that Before the second channel management interface receives the first channel identifier sent by the first channel management interface, the method further includes: The first channel management interface parses the slave address in the second channel request command; When the slave address of the second channel request command corresponds to the first channel management interface, the first channel management interface parses the command identifier of the second channel request command; When the command identifier of the second channel request command is a channel request command identifier, the first channel management interface generates the first channel identifier and sends it to the second channel management interface.

13. The method according to claim 12, characterized in that The first channel management interface generates the first channel identifier, including: The first channel management interface obtains the channel status between the management partition and the logical processing unit, where the channel status includes a channel identifier and the number of pending commands; The first channel management interface uses the channel identifier with the smallest number of commands to be processed as the first channel identifier.

14. The method according to claim 10, characterized in that Also includes: When receiving the response message, the service management interface of the management partition parses the slave address of the response message; When the slave address of the response message corresponds to the service management interface, the service management interface processes the response data.

15. The method according to claim 1, wherein Before the logical processing unit receives the first management command sent by the management partition, the method further includes: The first channel management interface of the management partition sends a first channel request command to the second channel management interface of the logical processing unit; The first channel management interface receives the second channel identifier sent by the second channel management interface; The management partition sends the first management command to the logical processing unit through the service management interface corresponding to the second channel identifier, and the slave address in the first management command corresponds to the first service processing interface corresponding to the second channel identifier in the logical processing unit.

16. The method according to claim 15, characterized in that Before the first channel management interface receives the second channel identifier sent by the second channel management interface, the method further includes: The second channel management interface parses the slave address in the first channel request command; When the slave address of the first channel request command corresponds to the second channel management interface, the second channel management interface parses the command identifier of the first channel request command; When the command identifier of the first channel request command is a channel request command identifier, the second channel management interface generates the second channel identifier and sends it to the first channel management interface.

17. The method according to claim 15, characterized in that The second channel management interface generates the second channel identifier, including: The second channel management interface parses the management partition identifier of the first channel request command and determines the second channel identifier according to the management partition identifier.

18. The method according to claim 17, characterized in that The determining the second channel identifier according to the management partition identifier includes: Acquire a channel status between the management partition corresponding to the management partition identifier and the logical processing unit, the channel status including the channel identifier and the number of pending commands; The channel identifier with the smallest number of commands to be processed is used as the second channel identifier.

19. The method according to claim 1, wherein The command identifier includes: a target device identifier and a management operation identifier.

20. A host partition management architecture, comprising: A host, a logic processing unit, and a management unit; the host includes multiple host partitions, and the management unit includes multiple management partitions; The logic processing unit is configured to execute the method according to any one of claims 1 to 19.

Images