CN102063341A - high density server - Google Patents

Abstract

The invention relates to a high-density server, which comprises a backboard, a bus and a plurality of main boards. The backboard comprises a storage component and a central management chip. Each motherboard comprises a basic input/output system, a baseboard management controller and a processing component. The processing component detects whether the basic input and output system on the same mainboard has errors. When the basic input and output system on the same mainboard has errors, the processing component requires the central management chip to acquire and return a basic input and output system mapping file from the storage component through the baseboard management controller, and performs error recovery on the basic input and output system according to the basic input and output system mapping file.

Description

high density server

technical field

The present invention relates to a high-density server, and in particular to a high-density server with error recovery function.

Background technique

With the rapid development of digital technology, the application of digital products is becoming more and more popular, and the progress of computer-related peripheral industries is even more rapid. With the popularization of computers and the rapid development of network technology, the network has become an important medium for fast access to data and can provide diversified information services. Therefore, due to the massive use of the network, network-related businesses must purchase a lot of network and computer-related equipment to fully meet the needs of users.

However, since the available space in the computer room is limited, it is desirable that the computer and related equipment be of a smaller size in order to reduce the office or factory space they occupy. Especially the large computer server system, in order to have high stability, so compared with the size of the general desktop computer, it is even worse.

Therefore, the development of the server focuses on providing higher processing performance per unit volume of the server and being able to be flexibly deployed in the server. Therefore, a high-density server (High Density server system) with multiple independent motherboards in a chassis has been developed.

Among the multiple motherboards of the high-density server, if an error occurs in the Basic Input/Output System (BIOS) of one of the motherboards, it is necessary to use the access device of the motherboard where the error occurred, Such as a floppy disk drive or a universal serial bus (Universal Serial Bus, USB) device, reads the BIOS image file (image) to perform BIOS error recovery. In this way, when a high-density server has a motherboard error, its administrator needs to go to the side of the high-density server to put the BIOS image file into the access device, and cannot remotely operate the BIOS error recovery.

Contents of the invention

Therefore, an object of the present invention is to provide a high-density server, which is used for when the BIOS of at least one of the plurality of motherboards of the high-density server has an error, according to the data stored in the backplane of the high-density server The BIOS image file of the storage component is used to perform error recovery on the error occurred in the BIOS.

A high-density server provided according to an embodiment of the present invention includes a backplane, a bus, and a processing component. The backplane includes a storage component and a central management chip (Enclosure Management controller, EM controller) electrically connected to each other. Each motherboard includes a BIOS, a Baseboard Management Controller (BMC) and a processing unit. The baseboard management controller is electrically connected to the central management chip through the bus. The processing component is electrically connected to the baseboard management controller and the BIOS. The processing component includes a fault detection module and a recovery module. The debug module detects whether there is an error in the BIOS of the same motherboard. When the BIOS of the same motherboard has an error, the recovery module requires the central management chip to obtain and return a BIOS image file from the storage component through its baseboard management controller. Next, the recovery module performs error recovery on the basic input and output system according to the basic input and output system image file.

According to an embodiment of the present invention, the central management chip includes a judging module for judging whether the BIOS image file exists in the storage component when receiving a request to obtain and return the BIOS image file. Wherein, when the BIOS image file is stored in the storage component, the judging module enables the central management chip to obtain and return the BIOS image file from the storage component.

According to another embodiment of the present invention, the central management chip further includes an update module, configured to receive another BIOS image file and update another BIOS image file to the storage when an update command is received. components.

According to another embodiment of the present invention, the backplane further includes a data transmission interface electrically connected to the central management chip. Wherein, the update command is received through the data transmission interface.

According to another embodiment of the present invention, the backplane further includes a network interface electrically connected to the central management chip for establishing a connection with the network. Wherein, the update command is received through the network.

According to another embodiment of the present invention, the bus is an Intelligent Platform Management Bus (IPMB).

According to another embodiment of the present invention, each of the processing components is electrically connected to the BMC of the same motherboard through an Intelligent Platform Management Interface (IPMI).

According to another embodiment of the present invention, the storage component is a flash memory (Flash) or a non-volatile memory (Non-Volatile Random Access Memory, NVRAM).

It can be seen from the above embodiments of the present invention that the application of the present invention has the following advantages. Among multiple main boards of a high-density server, if at least one main board's basic input and output system fails, each main board can automatically correct the wrong basic input and output system through the image file of the storage component stored on the backplane. The system responds with an error. In other words, the administrator of the high-density server can perform error recovery without personally providing the image file next to the high-density server. In addition, when errors occur in the BIOS of multiple mainboards, there is no need to individually provide image files for each mainboard for use in error recovery. In addition, by detecting whether there is an image file in the storage component, it is possible to avoid returning wrong data and causing error recovery failure.

Description of drawings

In order to make the above and other objects, features, advantages and embodiments of the present invention more comprehensible, the accompanying drawings are described as follows:

FIG. 1 is a functional block diagram of a high-density server according to an embodiment of the present invention.

[Description of main component symbols]

100: Backplane 310, 320,..., 3n0: Mainboard

110: storage components 311, 321, ..., 3n1: baseboard management controller

120: central management chip 312, 322, ..., 3n2: processing components

121: Judgment module 313, 323, ..., 3n3: Basic input and output system

122: Update module 314, 324, ..., 3n4: Debug module

130: Data transmission interface 315, 325, ..., 3n5: Reply module

140: Network interface 400: Network

200: bus

Detailed ways

Please refer to FIG. 1 , which shows a functional block diagram of a high-density server according to an embodiment of the present invention. When the BIOS of at least one of the plurality of motherboards of the high-density server has an error, according to the BIOS image file stored in the storage components of the backplane of the high-density server, the error occurs to the BIOS The system responds with an error.

The high-density server includes a backplane 100, a bus 200, and several motherboards 310, 320, . . . , 3n0. The backplane 100 includes a storage component 110 and a central management chip 120 electrically connected to each other. Wherein, the storage component 110 can be a flash memory, a non-volatile memory or other accessible storage components. Each motherboard 310, 320, . . . , 3n0 includes baseboard management controllers 311, 321, . ., 3n3. Each processing component 312, 322, . . . , 3n2 includes an error detection module 314, 324, . Wherein, the error detection modules 314, 324, . . . , 3n4 and the reply modules 315, 325, . communicate.

The baseboard management controllers 311 , 321 , . . . , 3n1 are electrically connected to the central management chip 120 through the bus 200 . Wherein, the bus 200 can be an intelligent platform management bus or other bus that can enable the central management chip 120 of the backplane 100 and the baseboard management controllers 311 , 321 , . . . , 3n1 to perform data transmission. Each processing component 312, 322, . . . , 3n2 is electrically connected to the baseboard management controller 311, 321, . Wherein, the processing components 312, 322, . The devices 311, 321, . . . , 3n1 are electrically connected.

Each error detection module 314, 324, . . . , 3n4 respectively detects whether there is an error in the BIOS 313, 323, . Wherein, the error detection modules 314, 324, . . . , 3n4 may perform error detection through checksum or other error detection methods.

When the BIOS 313, 323, . . . , 3n3 have errors, the recovery modules 315, 325, . , 3n1, requiring the central management chip 120 to obtain and return a BIOS image file from the storage component 110 . Then, the recovery modules 315, 325, . . . , 3n5 perform error recovery to the error-prone BIOS 313, 323, . . . , 3n3 according to the BIOS image file.

For example, when the error detection module 314 on the mainboard 310 detects that the BIOS 313 of the same mainboard 310 has an error, the recovery module 315 on the mainboard 310 passes through the baseboard management controller 311 of the mainboard 310 , requiring the central management chip 120 to obtain and return the BIOS image file from the storage component 110 . Then, the recovery module 315 of the motherboard 310 performs an error recovery to the BIOS 313 that has an error on the motherboard 310 according to the BIOS image file. However, in other embodiments, if the BIOS of other motherboards has errors, the recovery module on the same motherboard can also perform error recovery, which is not limited to this embodiment. In this way, when an error occurs in the BIOS of the motherboard, the error recovery can be performed through the image file stored in the storage component.

In order to prevent the central management chip 120 from returning wrong data when the storage component 110 does not store the BIOS image file, resulting in error recovery failure, the central management chip 120 may include a judging module 121 . When the central management chip 120 receives a request for acquiring and returning the BIOS image file, the judging module 121 judges whether the BIOS image file is stored in the storage component 110 . Wherein, when the BIOS image file is stored in the storage component 110, the judging module 121 enables the central management chip 110 to obtain and return the BIOS image file from the storage component to the requesting motherboard. In this way, it is possible to prevent the central management chip 120 from returning wrong data, resulting in failure of error reply.

In addition, the high-density server can update the BIOS stored in its storage component 110 . Therefore, the central management chip 120 may include an update module 123 for receiving another BIOS image file and updating the other BIOS image file to the storage component 110 when an update command is received. Wherein, another BIOS image file may be included in the update command, or the update command may include a storage location of another BIOS image file for the high-density server to obtain.

In addition, the backplane 100 may include a data transmission interface 130 or a network interface 140 electrically connected to the central management chip 120 . In this way, the update command can be received through the data transmission interface 130 , or establish a connection with the network 400 through the network interface, and receive the update command through the network 400 . In other words, the administrator of the high-density server can remotely issue an update command through the network to update the image file stored in the storage component.

It can be seen from the above embodiments of the present invention that the application of the present invention has the following advantages. Among multiple main boards of a high-density server, if at least one main board's basic input and output system fails, each main board can automatically correct the wrong basic input and output system through the image file of the storage component stored on the backplane. The system responds with an error. In other words, the administrator of the high-density server can perform error recovery without personally providing the image file next to the high-density server. In addition, when errors occur in the BIOS of multiple mainboards, there is no need to individually provide image files for each mainboard for use in error recovery. In addition, by detecting whether there is an image file in the storage component, it is possible to avoid returning wrong data and causing error recovery failure.

Although the present invention has been disclosed above in terms of implementation, it is not intended to limit the present invention. Any person familiar with this technology can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, this The protection scope of the invention should be determined by the scope defined in the claims.

Claims (8)

1. A high-density server, characterized in that it comprises: A backplane, comprising: a storage component; and a central management chip electrically connected to the storage element; a bus; and a plurality of motherboards, each of which includes: - basic input output system; a baseboard management controller electrically connected to the central management chip through the bus; and A processing component, electrically connected to the baseboard management controller and the BIOS, includes: an error detection module, which detects whether there is an error in the basic input output system; and A recovery module, when the BIOS has an error, through the baseboard management controller, requires the central management chip to obtain and return a BIOS image file from the storage component, and according to the BIOS Image file for error recovery on this BIOS. 2. The high-density server according to claim 1, wherein the central management chip comprises: A judging module, when receiving a request to obtain and return the BIOS image file, judge whether the BIOS image file is stored in the storage component, wherein the BIOS is stored in the storage component When the system image file is used, the judging module enables the central management chip to obtain and return the BIOS image file from the storage component. 3. The high-density server according to claim 1, wherein the central management chip further comprises: An update module receives another BIOS image file when receiving an update command, and updates the other BIOS image file to the storage component. 4. The high-density server according to claim 3, wherein the backplane further comprises: A data transmission interface is electrically connected to the central management chip, wherein the update command is received through the data transmission interface. 5. The high-density server according to claim 3, wherein the backplane further comprises: A network interface is electrically connected to the central management chip to establish a link with the network, wherein the update command is received through the network. 6. The high-density server according to claim 1, wherein the bus is an intelligent platform management bus. 7. The high-density server according to claim 1, wherein each of the processing components is electrically connected to its baseboard management controller through an intelligent platform management interface. 8. The high-density server according to claim 1, wherein the storage component is a flash memory or a non-volatile memory.

Images