TWI505674B - Server system and a data transferring method thereof - Google Patents

Description

Server system and its data transmission method

The present invention relates to a server system, and more particularly to a server system capable of improving bandwidth usage efficiency and a data transmission method thereof.

In recent years, with the rapid development of technology, the functions of computer systems have become more powerful. In order to effectively monitor the operation of various components on the motherboard, many motherboard manufacturers facilitate the use of Baseboard Management Control (BMC) to monitor various operations of the system and transmit the results of the monitoring to a management module.

In general, the baseboard management controller periodically polls different sensors on the motherboard to monitor the current working state of the hardware on the motherboard, and sends an inquiry to the baseboard management controller according to the management module. The message is transmitted to the management module for further processing. In other words, in the transmission mode, when the management module sends the inquiry message, the baseboard management controller does not simultaneously transmit the monitoring result to the management module; otherwise, when the baseboard management controller transmits the monitoring result to the management module, The management module will not send the inquiry message at the same time. Therefore, such a transmission method is not conducive to the efficiency of use of the bandwidth, so there is still room for further improvement.

In view of the inefficient use of the above bandwidth, the present invention automatically transmits the monitoring result to the management module according to a preset condition to solve the problem of bandwidth occupation.

One aspect of the present invention is to provide a server system. The server system includes a plurality of server nodes and at least one management module. Each of the server nodes includes a node control module, and the node control module can collect the running status information of the corresponding server node. Each node control module can communicate with the management module through an uplink data channel. Each node control module can encapsulate the running status information of the corresponding server node into a data packet, and automatically send the data packet to the management module along the uplink data channel according to a preset condition, and the management module receives and parses the data. The packet is managed, and the operation of the server node is managed according to the running status information of the corresponding server node.

In an embodiment, the preset condition is a fixed time period, and each time the fixed time period passes, each node control module sends the data packet to the management module along the uplink data channel.

In an embodiment, the preset condition is an information change. When the running status information of one of the server nodes changes, the node control module corresponding to the server node sends the data packet to the uplink data channel. At least one management module.

In an embodiment, the management module further includes an error correction module to detect the servo when one of the server nodes has an information push error. And the error correction module determines whether the information push error occurs by detecting whether the node control module corresponding to one of the server nodes stops the information push action.

In an embodiment, the data channel is further included, and the management module sends a command to the server nodes through the downlink data channel, wherein the command is a power on/off command.

In an embodiment, the node control module can encapsulate the running state information of the corresponding server node into a plurality of data packets, and a system management controller is coupled to the management module, and the management module receives the plurality of data modules. The data packet is secondarily packaged and sent to the system management controller, wherein the running status information is temperature information.

In an embodiment, the system management controller can be a Racks Management Controller (RMC), the management module can be a Fan Controller Board (FCB), and the node control module can be a A baseboard management controller (BMC), wherein the fan management control board is further electrically connected to a fan module for dissipating heat from the server nodes, and adjusts the speed of each fan in the fan module according to the temperature information.

In one embodiment, the management module is a cabinet management controller, the node control module is a baseboard management controller, and each server node is further connected to a fan, and the baseboard management controller controls the fan according to the temperature information. Rotating speed.

Another aspect of the present invention is to provide a server system data transfer method. The server system includes at least a plurality of server nodes and at least one management module. First, the running status information of the server nodes is collected, and the running status information is encapsulated into a plurality of data packets. Then, based on a predetermined condition The data packets are automatically sent to the management module along an uplink data channel. Finally, the data packets are received and parsed, wherein the management module manages the running of the server nodes according to the running status information of the server nodes.

In summary, the present invention allows the node control module to actively transmit the running status information of the corresponding server to the management module according to a preset condition, and the management module does not need to send an inquiry message to the node control module. Therefore, it does not occupy the bandwidth of sending the inquiry message, and is more efficient in bandwidth usage.

100 and 200 and 300‧‧‧ server systems

101‧‧‧ server node

102‧‧‧Management module

103‧‧‧Upstream data channel

104‧‧‧Down data channel

105‧‧‧System Management Controller

1011‧‧‧node control module

1012‧‧‧Server

1021 and 2021 and 3011‧‧‧ error correction modules

201 and 301‧‧‧Cabinet Management Controller

202‧‧‧Fan Management Control Board

203 and 303‧‧‧Base Management Controller

204 and 304‧‧‧ motherboards

205‧‧‧Fan module

2041 and 3041‧‧‧Operation Processing Unit

3042‧‧‧fan

2043 and 3043‧‧‧ Temperature Sensors

401 and 402 and 403‧‧ steps

Figure 1 is a schematic illustration of a server system in accordance with an embodiment of the present invention.

Figure 2 is a schematic illustration of a server system in accordance with another embodiment of the present invention.

Figure 3 is a schematic illustration of a server system in accordance with yet another embodiment of the present invention.

FIG. 4 is a diagram showing a data transmission method of a server system according to an embodiment of the present invention.

The following description of the preferred embodiments of the invention is in the

Figure 1 is a schematic illustration of a server system in accordance with an embodiment of the present invention. The server system 100 of the present invention includes: a plurality of server nodes 101 to And at least one management module 102. Each of the server nodes 101 further includes a node control module 1011 and a corresponding server 1012. The node control module 1011 can collect the running status information of the corresponding server 1012, that is, the node. In addition, the node control module 1011 can be coupled to the management module 102 through an uplink data channel 103, so as to transmit the running status information of the server 1012 to the management module 102. Traditionally, after the management module 102 first sends a query message to the node control module 1011, the node control module 1011 transmits the running status information of the corresponding server 1012 to the management module 102. The transmission mode is not conducive to the efficiency of bandwidth usage. Therefore, the present invention allows the node control module 1011 to transmit the running status information of the corresponding server 1012 to the management module 102 according to a preset condition, that is, as long as the preset condition is satisfied, the node control module 1011 automatically responds. The running status information of the server 1012 is automatically transmitted to the management module 102. The management module 102 does not need to send an inquiry message to the node control module 1011 first. Therefore, the bandwidth is not occupied and the bandwidth is more efficient.

The node control module 1011 periodically polls different sensors (not shown) on the corresponding server 1012 to monitor the current running state of the server 1012, and the corresponding server 1012. The operational status information classification is encapsulated into multiple data packets. When the preset condition is met, the node control module 1011 will actively send the data packet to the management module 102 along the uplink data channel 103. After the management module 102 receives the data packet uploaded by the node control module 1011, the data packet is parsed, and the running of each server 1012 is managed according to the running status information contained in the data packet. In an embodiment, the preset condition is a fixed time period, that is, the node control module 1011 actively sends the current running state of the server 1012 along the uplink data channel 103 every time a fixed time period elapses. give Management module 102. In another embodiment, the preset condition is an information change, for example, a change in the running state of the server 1012, that is, whenever the node control module 1011 detects that the current running state of the corresponding server 1012 occurs. A drastic change, for example, cannot be operated. At this time, the node control module 1011 actively sends the current running state of the server 1012 to the management module 102 along the uplink data channel 103. In practice, a threshold may be set. When the rate of change of the operating state exceeds the threshold, the node control module 1011 actively sends the current running state of the server 1012 to the management module 102. For example, the threshold value is set to 2 ° C. When the node control module 1011 detects that the current temperature change of the corresponding server 1012 exceeds 2 ° C, the node control module 1011 actively sends the current running status of the server 1012 to the management module 102. . In addition, the management module 102 of the present invention further includes an error correction module 1021, wherein when one of the server nodes 101 has an information push error, the error correction module 1021 can detect the error servo. Node. In an embodiment, the error correction module 1021 determines whether a information push error occurs by detecting whether the node control module 1011 corresponding to each server node 101 stops the information push operation. The server system 100 further includes a data channel 104. The management module 102 can send commands to the server nodes through the downlink data channel 104. For example, the command node control module 1011 controls the corresponding server 1012 to be powered on or off.

In addition, the server system of the present invention further includes a system management controller 105 coupled to the management module 102. If the server system has multiple management modules 102, the system management controller 105 can be coupled to the plurality of management modules 102 for integrated management control of the entire server system. Each management module 102 uploads the node control module 1011 and classifies and encapsulates multiple data packets for secondary sealing. Installed and uploaded to the system management controller 105. The system management controller 105 can determine the information packet uploaded by the management module 102 according to the encapsulated data packet of the category, so as to manage the entire server system. For example, in an embodiment, the corresponding server 1012 includes a motherboard and a cooling fan, and the node control module 1011 periodically polls different sensors on the corresponding server 1012 to separately monitor the motherboard and the cooling fan. The current running state, and the operating state information of the motherboard and the cooling fan are packaged into data packets and transmitted to the management module 102. The management module 102 repackages the data package of the motherboard and the data package of the cooling fan, and includes information corresponding to the management module 102, and then uploads the information to the system management controller 105. The system management controller 105 can determine which management module 102 uploads the operating status information according to the information after the secondary encapsulation, and performs corresponding management.

In an embodiment, the system management controller 105 shown in FIG. 1 is, for example, a Racks Management Controller (RMC). The management module 102 is, for example, a fan controller board (FCB). The node control module 1011 is, for example, a Baseboard Management Controller (BMC) provided on the motherboard. Figure 2 is a schematic illustration of a server system in accordance with this embodiment of the invention. The server system 200 includes a cabinet management controller 201, a fan management control board 202, a substrate management controller 203, and a fan module 205 having a plurality of fans. The substrate management controller 203 is disposed on a motherboard 204. The motherboard 204 includes, for example, an arithmetic processing unit 2041 and a temperature sensor 2043. The fan management control board 202 is electrically connected to the fan module 205, and the fan module 205 is configured to dissipate heat from the plurality of server nodes. The fan management control board 202 is configured according to an information, for example, on each motherboard. The temperature information of the processing unit 2041 adjusts the rotation speed of the corresponding fan in the fan module 205, thereby controlling the operating temperature of the corresponding operation processing unit 2041. It should be noted that in the present embodiment, the application of the present invention is described only by the operation processing unit 2041 provided on the motherboard 204. However, the application and structure of the present invention are not limited to the above embodiments. According to the embodiment, the substrate management controller 203 repeatedly reads the measured operating temperature value in a polling manner for the temperature sensor 2043 of each motherboard 204 to obtain the operating temperature value of the arithmetic processing unit 2041. And measuring the measured operating temperature value into a data packet.

When the preset condition is met, the baseboard management controller 203 actively sends the data packet to the fan management control board 202. After the fan management control board 202 receives the data packet uploaded by the substrate management controller 203, the data packet is parsed, and the operating state of the fan module 205 is controlled according to the operating temperature value of the arithmetic processing unit 2041 contained in the data packet. In an embodiment, the preset condition is a fixed time period, that is, the substrate management controller 203 actively encapsulates the operating temperature value of the operation processing unit 2041 into a data packet transmission every time a fixed time period elapses. The fan management control board 202 is controlled to perform the operation state of the fan module 205. In another embodiment, the preset condition is an information change, for example, a change in the operating state, that is, each time the baseboard management controller 203 monitors the running state of the operation processing unit 2041 on a motherboard. A drastic change occurs and exceeds a threshold. For example, if the sudden temperature rise exceeds the set threshold, the baseboard management controller 203 will actively send the burst status to the fan management control board 202 for immediate processing. In an embodiment, the threshold value is set to 2 ° C. When the substrate management controller 203 monitors that the current temperature change of the arithmetic processing unit 2041 exceeds 2 ° C, the substrate management controller 203 actively takes the current operation of the arithmetic processing unit 2041. The row status is sent to the fan management control board 202 for corresponding processing, such as increasing the operating speed of the fan module 205, so as to instantly cool down.

On the other hand, the fan management control board 202 is further coupled to a cabinet management controller 201. The fan management control board 202 performs secondary encapsulation on the operating temperature value of the operation processing unit 2041 uploaded by the baseboard management controller 203 to include information corresponding to the fan management control board 202, and then uploads the information to the rack management controller 201. The cabinet management controller 201 can determine the information packet uploaded by the fan management control board 202 according to the information package encapsulated in the category, so as to perform corresponding management. In addition, the fan management control board 202 further includes an error correction module 2021, so that when the substrate management controller 203 uploads the fan management control board 202 data packet occurrence information push error, the faulty substrate management controller 203 is detected. The error correction module 2021 determines whether or not an information push error occurs by detecting whether each of the substrate management controllers 203 stops the information push operation.

In another embodiment, the management module 102 shown in FIG. 1 is, for example, a Racks Management Controller (RMC). The node control module 1011 is, for example, a Baseboard Management Controller (BMC) provided on the motherboard. Figure 3 is a schematic illustration of a server system in accordance with this embodiment of the present invention. The server system 300 includes a cabinet management controller 301 and a baseboard management controller 303. The substrate management controller 303 is disposed on a motherboard 304. The motherboard 304 includes, for example, an arithmetic processing unit 3041, a fan 3042, and a temperature sensor 3043. It should be noted that in the present embodiment, the application processing unit 3041 and the fan 3042 are provided on the motherboard 304 to explain the application of the present invention. However, the application and structure of the present invention are not limited to the above embodiments. According to the embodiment, the substrate management controller 303 is based on an information, for example, The temperature information of the arithmetic processing unit 3041 on each motherboard adjusts the rotation speed of the corresponding fan 3042, thereby controlling the operating temperature of the arithmetic processing unit 3041. The substrate management controller 303 repeatedly reads the measured operating temperature value of the arithmetic processing unit 3041 in a polling manner for the temperature sensor 3043 of each motherboard 304, and adjusts the fan 3042 according to the temperature information. The speed of rotation. When the preset condition is met, the baseboard management controller 303 actively sends the data packet to the rack management controller 301 for subsequent processing. In an embodiment, the preset condition is a fixed time period, that is, the substrate management controller 303 actively takes the operation processing unit 3041 operating temperature value and the operating state of the fan 3042 every time a fixed time period elapses. The classification is encapsulated into data packets and sent to the cabinet management controller 301 for subsequent control. In addition, the rack management controller 301 further includes an error correction module 3011 for detecting the faulty substrate management controller 303 when the substrate management controller 303 uploads the data packet to the rack management controller 301 for an information push error. The error correction module 3011 determines whether the substrate management controller 303 has an information push error by detecting whether each of the substrate management controllers 303 stops the information push operation.

FIG. 4 is a diagram showing a data transmission method of a server system according to an embodiment of the present invention. Please refer to FIG. 1 and FIG. 4 at the same time. First, in step 401, the running state information of the server node is collected, and the running state information is encapsulated into a plurality of data packets. In an embodiment, the server system includes at least a plurality of server nodes and at least one management module. Each server node 101 further includes a node control module 1011 and a corresponding server 1012. The node control module 1011 The corresponding server 1012, that is, the running status information of the node, can be collected and encapsulated into a plurality of data packets. Then, in step 402, the automatic conditions are automatically determined according to a preset condition. The data packet is sent to the management module along an uplink data channel. In an embodiment, the node control module 1011 is coupled to the management module 102 via an uplink data channel 103, and transmits a plurality of data packets to the management module 102 according to a preset condition, wherein the preset condition is, for example, A fixed time period or an operating state information change, that is, each time a fixed time period passes or the operating state information changes, the node control module 1011 actively sends the current running state of the server 1012 along the uplink data channel 103. Management module 102. Finally, in step 403, the management module receives and parses the data packet to manage the operation of the server nodes according to the running status information of the server nodes.

In summary, the present invention allows the node control module to actively transmit the running status information of the corresponding server to the management module by setting a preset condition, and the management module does not need to send an inquiry message to the node control module. Therefore, it does not occupy the bandwidth of sending the inquiry message, and only occupies the bandwidth when transmitting the server running status information to the management module, so it is more efficient in bandwidth usage.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the present invention, and the present invention can be modified and modified without departing from the spirit and scope of the present invention. The scope is subject to the definition of the scope of the patent application attached.

100‧‧‧Server system

101‧‧‧ server node

102‧‧‧Management module

103‧‧‧Upstream data channel

104‧‧‧Down data channel

105‧‧‧System Management Controller

1011‧‧‧node control module

1012‧‧‧Server

1021‧‧‧Error Correction Module

Claims (24)

A server system includes at least: a plurality of server nodes, each server node including a node control module, wherein the node control module can collect operating state information of a corresponding server node; and at least one management module, Each of the node control modules is respectively connected to the at least one management module through an uplink data channel; wherein each of the node control modules encapsulates the running status information of the corresponding server node into a data packet, and The data packet is automatically sent to the at least one management module along the uplink data channel according to a preset condition, and the at least one management module receives and parses the data packet, and manages the data packet according to the running state information of the corresponding server node. The operation of the server node. The server system of claim 1, wherein the preset condition is a fixed time period, wherein each of the node control modules encapsulates the data packet along the uplink data channel every time the fixed time period elapses Send to the at least one management module. The server system of claim 1, wherein the preset condition is an information change, and when the operating state information of one of the server nodes changes, the node control mode corresponding to the server node The group sends the data packet to the at least one management along the uplink data channel Module. The server system of claim 1, wherein the management module further comprises an error correction module for detecting the server node when an information push error occurs in one of the server nodes. The server system of claim 4, wherein the error correction module determines whether the information push is triggered by detecting whether the node control module corresponding to one of the server nodes stops the information push action error. For example, the server system described in claim 1 further includes a data channel, and the management module sends a command to the server nodes through the downlink data channel. The server system of claim 6, wherein the command is a power on/off command. The server system of claim 1, wherein each of the node control modules encapsulates the operating state information of the corresponding server node into a plurality of data packets. The server system of claim 8, further comprising a system management controller coupled to the at least one management module, the at least one tube The module performs secondary encapsulation on the received plurality of data packets and sends them to the system management controller. The server system of claim 9, wherein the system management controller is a Racks Management Controller (RMC), and the management module is a fan management control board (Fan Controller Board, FCB). The node control module is a Baseboard Management Controller (BMC). The server system of claim 10, wherein the operating status information is temperature information. The server system of claim 11, wherein the at least one fan management control board is further electrically connected to a fan module for dissipating heat to the plurality of server nodes, and adjusting the fan according to the temperature information. The speed of each fan in the module. The server system of claim 1, wherein the management module is a cabinet management controller, and the node control module is a baseboard management controller. The server system of claim 13, wherein the operating status information is temperature information. The server system of claim 14, wherein each of the server nodes is further connected to a fan, and the baseboard management controller controls the rotation speed of the fans according to the temperature information. A server system data transmission method, wherein the server system includes at least a plurality of server nodes and at least one management module, the method comprising: collecting operation state information of the server nodes, and packaging the operation state information Forming a plurality of data packets; automatically transmitting the data packets to the at least one management module along an uplink data channel according to a preset condition; and receiving and parsing the data packets, wherein the at least one management module is configured according to the The running status information of the server node manages the operation of the server nodes. The server system data transmission method according to claim 16, wherein the preset condition is a fixed time period, and the data packets are automatically sent to the at least one management along the uplink data channel according to the fixed time period. Module. The method for transmitting data of a server system according to claim 16, wherein the preset condition is an information change, and when the running status information of one of the server nodes changes, the corresponding data is automatically encapsulated. And sending the at least one management module along the uplink data channel. The server system data transmission method as described in claim 16 further includes: correcting the server nodes to detect the servo when an information push error occurs in one of the server nodes. Node. The method for transmitting a server system data according to claim 19, further comprising: detecting whether one of the server nodes stops the information pushing action to determine whether a information pushing error occurs. The server system data transmission method of claim 16, further comprising: sending, by the at least one management module, a command to the server nodes by using a downlink data channel. The server system data transmission method according to claim 21, wherein the command is a power on/off command. For example, the server system data transmission method described in claim 16 further includes: classifying and packaging the running status information of the server nodes into a plurality of data packets. The method for transmitting a server system data according to claim 23, further comprising: sub-packaging the plurality of data packets and sending the data to a system management controller.