Apparatus and method for comparing heat dissipation capabilities of components in a server

Abstract

This invention belongs to the field of server heat dissipation testing technology, specifically providing an apparatus and method for comparing the heat dissipation capabilities of components in a server. The apparatus includes a main controller and a temperature chamber controlled by the main controller. At least two identical servers are installed inside the temperature chamber. Components to be compared are installed at the same location on each server. Temperature sensors for detecting the temperature of the corresponding component are installed at the location of the component to be compared on each server, and the temperature sensors are communicatively connected to the main controller. Each server contains a stress test program for the component to be compared. The main controller controls each server to power on and run the stress test program. The temperature detected by the temperature sensors is acquired in real time, and the component with the highest temperature in each test group is recorded as the benchmark component. After all components at the same location have been tested, the record information of the benchmark component with the highest temperature is output, quickly and accurately identifying the component with the worst heat dissipation capability among components of the same type.

Description

Technical Field

[0001] This invention relates to the field of server heat dissipation testing technology, and specifically to an apparatus and method for comparing the heat dissipation capabilities of components in a server. Background Technology

[0002] Thermal design and testing are crucial aspects of server development. Insufficient heat dissipation directly impacts the lifespan and reliability of servers and other computer products. Currently, the general approach is to compare the power consumption of similar components against their specifications, selecting the component with the highest power consumption for thermal development and testing. For example, if a list includes 10 similar hard drives, the power consumption values ​​are typically compared, and the component with the highest power consumption is chosen for thermal development and testing. This is because, generally, the higher the power consumption of a component, the more heat it generates, the worse its heat dissipation, and the more prone it is to overheating.

[0003] However, judging a component's heat dissipation capability solely based on its power consumption in the specifications carries certain risks. Higher power consumption doesn't necessarily mean poorer heat dissipation. Therefore, in server development, how to conduct effective server thermal development testing—that is, how to efficiently identify a component with poor heat dissipation capabilities for thermal development testing—is a problem that needs to be solved. Summary of the Invention

[0004] Judging a component's heat dissipation capability solely based on its power consumption in the specifications carries certain risks. Higher power consumption does not necessarily mean poorer heat dissipation. Therefore, in server development, how to conduct effective server heat dissipation development testing—that is, how to efficiently identify a component with poor heat dissipation capabilities for heat dissipation development testing—is a problem that needs to be solved. This invention provides an apparatus and method for comparing the heat dissipation capabilities of components in a server.

[0005] In a first aspect, the present invention provides a device for comparing the heat dissipation capacity of components in a server, including a main controller and a temperature chamber whose temperature is controlled by the main controller.

[0006] At least two servers with identical configurations should be installed inside the incubator;

[0007] The components to be compared are installed at the same location on each server; a temperature sensor is installed at the location of the component to be compared on each server to detect the temperature of the corresponding component, and the temperature sensor is connected to the main controller.

[0008] Each server contains a stress test program for the components to be compared.

[0009] The main controller controls each server to start up and run the stress test program; and acquires the temperature detected by the temperature sensor in real time, records the component with the highest temperature in each test group as the benchmark component, and records the identification code and benchmark temperature of the benchmark component. After all components at the same location have been tested, the recorded information of the benchmark component with the highest temperature is output; among them, components installed at the same location on the server and subjected to stress test at the same time are considered as a group.

[0010] Preferably, the device is further equipped with a switch, and the main controller is connected to the switch via a network cable;

[0011] The switch is connected to the network ports of each server inside the incubator.

[0012] Preferably, two servers are installed inside the temperature chamber; all components to be compared are installed and tested in pairs at the same location on the servers, and the record information of the component with the highest temperature is output.

[0013] Preferably, the main controller is used to select the component with the highest temperature as the reference component during the first set of tests, and the corresponding temperature is the reference temperature. During the next set of tests, the acquired temperature is compared with the reference temperature, and the highest temperature is selected and reset as the reference temperature, with the corresponding component becoming the reference component.

[0014] Preferably, the main controller is also used to determine whether the temperature difference collected by the same temperature sensor before and after the pressure test reaches the set time is within the set threshold range. If so, compare the temperatures of the same group, record the component with the highest temperature in each group of tests as the reference component, and record the identification code and reference temperature of the reference component. If not, wait for the set time again to determine the temperature difference collected by the same temperature sensor.

[0015] Secondly, the present invention also provides a method for comparing the heat dissipation capabilities of components in a server. This method is based on the apparatus described in the first aspect and includes the following steps:

[0016] Install the components to be compared in the same location on the server and connect the temperature sensors.

[0017] The main controller sets the temperature of the chamber via signal lines;

[0018] The main controller controls the server to power on and run the stress test program for the components via a network cable.

[0019] Temperature sensors acquire the temperature of components in each server and transmit it to the main controller;

[0020] The main controller obtains the highest temperature of the component at the same location on each server and sets it as the reference temperature, and sets the corresponding component as the reference component; and records the identification code and reference temperature of the reference component;

[0021] Output the recorded information of the reference components at all locations to be tested on the server.

[0022] Preferably, after the step of the temperature sensor acquiring the temperature of the components in each server and transmitting it to the main controller, the following steps are included:

[0023] The main controller determines whether the reference temperature exists;

[0024] If so, compare the temperature of each component at the same location on the server with the reference temperature. If the temperature of a component is higher than the reference temperature, execute the following steps: The main controller obtains the highest temperature of each component at the same location on the server and sets it as the reference temperature, and sets the corresponding component as the reference component; and records the identification code and reference temperature of the reference component.

[0025] If not, keep the original reference temperature and corresponding reference components unchanged.

[0026] Preferably, after the main controller obtains the highest temperature of the component at the same location on each server and sets it as the reference temperature, and the corresponding component is set as the reference component; and records the identification code and reference temperature of the reference component, the following steps are included:

[0027] Determine if all components in the same location on the server have completed testing;

[0028] If so, output the record information of the benchmark component in the same location of the server to the test list;

[0029] If not, reselect the component to be compared at that location and perform the following steps: install the components to be compared in the same location on the server and connect the temperature sensor.

[0030] Preferably, after the step of outputting the record information of the reference component in the same location component of the server to the test list, the following steps are included:

[0031] Determine whether the testing of all components to be compared at all test locations has been completed;

[0032] If so, the following steps are executed: Output the record information of the reference components at all test locations on the server, that is, output the test list;

[0033] If not, proceed as follows: Install the components to be compared in the same location on the server and connect the temperature sensors.

[0034] As can be seen from the above technical solutions, the present invention has the following advantages: This device and method can quickly and accurately identify the component with the worst heat dissipation capacity among similar components. Then, the component with the worst heat dissipation is used in the development and testing of servers, which can improve the quality of server heat dissipation development and testing.

[0035] Furthermore, the design principle of this invention is reliable, the structure is simple, and it has a very wide range of application prospects.

[0036] Therefore, it is evident that the present invention has outstanding substantive features and significant progress compared with the prior art, and the beneficial effects of its implementation are also obvious. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a schematic flowchart of a method according to an embodiment of the present invention.

[0039] Figure 2 This is a schematic block diagram of an apparatus according to an embodiment of the present invention. Detailed Implementation

[0040] Currently, the general approach is to select components with higher power consumption from similar components by comparing their specifications for thermal development and testing. For example, if a list contains 10 similar hard drives, the power consumption values ​​are typically compared in their specifications, and the component with the highest power consumption is selected for thermal development and testing. This is because, generally, the higher the power consumption of a component, the more heat it generates, the worse its heat dissipation, and the more prone it is to overheating. However, judging a component's thermal performance solely based on its power consumption in the specifications carries certain risks. Higher power consumption does not necessarily mean worse heat dissipation. Therefore, in server development, how to conduct effective server thermal development and testing, i.e., how to efficiently find a component with poor heat dissipation for thermal development and testing, is a problem that needs to be solved. To enable those skilled in the art to better understand the technical solutions in this invention, the technical solutions in the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0041] like Figure 1 As shown, this embodiment of the invention provides a device for comparing the heat dissipation capabilities of components in a server, including a main controller and a temperature chamber whose temperature is controlled by the main controller.

[0042] At least two servers with identical configurations should be installed inside the incubator;

[0043] The components to be compared are installed at the same location on each server; a temperature sensor is installed at the location of the component to be compared on each server to detect the temperature of the corresponding component, and the temperature sensor is connected to the main controller.

[0044] Each server contains a stress test program for the components to be compared.

[0045] The main controller controls each server to power on and run the stress test program; it also acquires the temperature detected by the temperature sensor in real time, records the component with the highest temperature in each test group as the benchmark component, and records the identification code and benchmark temperature of the benchmark component. After all components at the same location have been tested, the recorded information of the benchmark component with the highest temperature is output. Components installed at the same location on the server and undergoing stress testing simultaneously constitute one group. The main controller is also used to determine whether the temperature difference collected by the same temperature sensor before and after the set time has elapsed during the stress test is within a set threshold range. If so, it compares the temperatures within the same group, records the component with the highest temperature in each test group as the benchmark component, and records the identification code and benchmark temperature of the benchmark component; if not, it waits for the set time again to determine the temperature difference collected by the same temperature sensor.

[0046] To improve testing efficiency, at least two servers are set up inside the chamber. The main controller needs to connect to the two servers through a switch. Therefore, the device is also equipped with a switch. The main controller is connected to the switch via a network cable. The switch is connected to the network ports of each server inside the chamber.

[0047] The testing process is as follows:

[0048] The components to be compared are installed in the same location on the server and connected to temperature sensors. The main controller sets the temperature of the chamber via signal lines. The main controller controls the server to start and run the stress test program for the components via network cables. The temperature sensors acquire the temperature of the components in each server and transmit it to the main controller. The main controller acquires the highest temperature of the components at the same location on each server and sets it as the reference temperature, and sets the corresponding component as the reference component. The main controller records the identification code and reference temperature of the reference component. The recorded information of the reference components at all test locations on the server is output.

[0049] In some embodiments, during testing, the main controller determines whether a reference temperature exists. If so, it compares the temperatures of each component at the same location on the server with the reference temperature. If a component's temperature is higher than the reference temperature, the following steps are executed: the main controller acquires the highest temperature of each component at the same location on the server and sets it as the reference temperature, designating the corresponding component as the reference component; and records the identification code and reference temperature of the reference component. If not, the original reference temperature and corresponding reference component remain unchanged. Further testing requires determining whether all components at the same location on the server have been tested. If so, the recorded information of the reference components at the same location on the server is output to the test list. If not, the component to be compared at that location is reselected, and the following steps are executed: the components to be compared are installed at the same location on the server, and temperature sensors are connected.

[0050] Determine whether the testing of all components to be compared at all test locations has been completed; if yes, execute the following steps: output the record information of the reference components at all test locations on the server, that is, output the test list; if no, execute the following steps: install the components to be compared at the same location on the server and connect the temperature sensor.

[0051] In some embodiments, two servers are installed inside the temperature chamber. All components to be compared are installed and tested in pairs at the same location on the servers, and the recording information of the component with the highest temperature is output. The main controller is used to select the component with the highest temperature as the reference component during the first test group, and the corresponding temperature is set as the reference temperature. During the next test group, the acquired temperature is compared with the reference temperature, and the highest temperature is selected and reset as the reference temperature, with the corresponding component becoming the reference component. During the testing process, components A and B to be compared are placed at the same location on both servers, and temperature sensors are connected. The main controller controls the temperature chamber to set a fixed temperature via a signal line, and then controls the two servers to power on and enter the system via a network cable, automatically running the pre-set stress test program for components A and B. Temperature sensors are set up to monitor the temperatures of components A and B, and transmit the data to the main controller for analysis and comparison in real time. After the set stress test time, the temperatures of components A and B will tend to stabilize. At this time, a comparison is made between the two temperatures a and b and the original reference temperature n. The component with the higher temperature is defined as the reference component, and its corresponding temperature is recorded in the main controller as the reference temperature n for that type of component.

[0052] When new components C and D of the same type are introduced, they are placed in their corresponding positions on both servers. The testing is repeated, and finally, temperatures c and d are generated and compared with a baseline temperature n. The highest temperature is designated as the baseline temperature, and the corresponding component becomes the baseline component. (That is, if the temperature of C or D is higher than N, it replaces N as the new baseline component; if it is lower than N, the original baseline temperature and component remain unchanged). After all components at the same location have been tested, N = max(a, b, n) is output, and the corresponding component becomes the baseline component.

[0053] This invention also provides a method for comparing the heat dissipation capabilities of components in a server. The method is based on a device for comparing the heat dissipation capabilities of components in a server. The device includes a main controller and a temperature chamber controlled by the main controller. At least two identical servers are installed inside the temperature chamber. Components to be compared are installed at the same location on each server. A temperature sensor for detecting the temperature of the corresponding component is installed at the location of the component to be compared on each server, and the temperature sensor is communicatively connected to the main controller. Each server contains a stress test program for the component to be compared. The method includes the following steps:

[0054] Step 1: Install the components to be compared in the same location on the server and connect the temperature sensors;

[0055] Step 2: The main controller sets the temperature of the chamber via signal lines;

[0056] Step 3: The main controller controls the server to power on and run the stress test program for the components via a network cable;

[0057] Step 4: Temperature sensors acquire the temperature of components in each server and transmit it to the main controller;

[0058] Step 5: The main controller obtains the highest temperature of the component at the same location on each server and sets it as the reference temperature, and sets the corresponding component as the reference component; and records the identification code and reference temperature of the reference component;

[0059] Step 6: Output the recorded information of the reference components at all locations to be tested on the server.

[0060] In some embodiments, after the step of the temperature sensor acquiring the temperature of the components in each server and transmitting it to the main controller, the following steps are included:

[0061] The main controller determines whether the reference temperature exists;

[0062] If so, compare the temperature of each component at the same location on the server with the reference temperature. If the temperature of a component is higher than the reference temperature, execute the following steps: The main controller obtains the highest temperature of each component at the same location on the server and sets it as the reference temperature, and sets the corresponding component as the reference component; and records the identification code and reference temperature of the reference component.

[0063] If not, keep the original reference temperature and corresponding reference components unchanged.

[0064] In some embodiments, after the main controller obtains the highest temperature of a component at the same location on each server and sets it as a reference temperature, and the corresponding component is set as a reference component; and records the identification code of the reference component and the reference temperature, the following steps are included:

[0065] Determine if all components in the same location on the server have completed testing;

[0066] If so, output the record information of the benchmark component in the same location of the server to the test list;

[0067] If not, reselect the component to be compared at that location and perform the following steps: install the components to be compared in the same location on the server and connect the temperature sensor.

[0068] In some embodiments, the step of outputting the record information of the reference component in the same location component of the server to the test list includes:

[0069] Determine whether the testing of all components to be compared at all test locations has been completed;

[0070] If so, the following steps are executed: Output the record information of the reference components at all test locations on the server, that is, output the test list;

[0071] If not, proceed as follows: Install the components to be compared in the same location on the server and connect the temperature sensors.

[0072] like Figure 2As shown, two servers are set up and tested, requiring identical configurations. Then, components A and B to be compared are placed in the same positions on both servers, and temperature sensors are connected. The main control unit controls the temperature chamber to set a fixed temperature via signal line 1, and then controls both servers to power on and automatically run the pre-set pressure programs for components A and B via network cable. The temperature sensors monitor the temperatures of components A and B and transmit the data to the main control unit for analysis and comparison in real time. After a certain period of pressure testing, the temperatures of components A and B will stabilize. At this point, a comparison is made between the two temperatures a and b and the original reference temperature n. The component with the higher temperature is defined as the reference component, and its corresponding temperature is recorded in the main control unit as the reference temperature n for that type of component. When new components C and D of the same type are introduced, they are similarly placed in the corresponding positions on the two servers. The test is repeated, and finally, temperatures c and d are generated and compared with the reference temperature n. The highest temperature is the reference temperature, and the corresponding component is the reference component. (That is, if the temperature of C or D is higher than that of N, then it replaces N as the new reference component; if it is lower than that of N, then the original reference temperature and reference component remain unchanged.) Later, when conducting server heat dissipation tests, the corresponding component can be found in the list by searching for the identification code, making testing convenient.

[0073] Although the present invention has been described in detail with reference to the accompanying drawings and preferred embodiments, the invention is not limited thereto. Various equivalent modifications or substitutions can be made to the embodiments of the invention by those skilled in the art without departing from the spirit and essence of the invention, and such modifications or substitutions should all be within the scope of the invention. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the invention should also be covered within the protection scope of the invention. Therefore, the protection scope of the invention should be determined by the scope of the claims.

Claims

1. A device for comparing the heat dissipation capabilities of components in a server, characterized in that, Includes a main controller and a temperature chamber whose temperature is controlled by the main controller; At least two servers with identical configurations should be installed inside the incubator; The components to be compared are installed at the same location on each server; a temperature sensor is installed at the location of the component to be compared on each server to detect the temperature of the corresponding component, and the temperature sensor is connected to the main controller. Each server contains a stress test program for the components to be compared. The main controller controls each server to power on and run the stress test program; and acquires the temperature detected by the temperature sensor in real time, records the component with the highest temperature in each test group as the benchmark component, and records the identification code and benchmark temperature of the benchmark component. After all components at the same location have been tested, the recorded information of the benchmark component with the highest temperature is output; among them, components installed at the same location on the server and undergoing stress testing at the same time are considered as a group. The main controller is used to select the component with the highest temperature as the reference component during the first set of tests, and the corresponding temperature is the reference temperature. During the next set of tests, the acquired temperature is compared with the reference temperature, and the highest temperature is selected and reset as the reference temperature, and the corresponding component is the reference component. The main controller is also used to determine whether the temperature difference collected by the same temperature sensor before and after the pressure test reaches the set time is within the set threshold range. If so, it compares the temperatures of the same group, records the component with the highest temperature in each group of tests as the reference component, and records the identification code and reference temperature of the reference component. If not, it waits for the set time again to determine the temperature difference collected by the same temperature sensor. By comparing the heat dissipation capabilities of components in a server, the device can quickly and accurately identify the component with the worst heat dissipation capability among similar components, and use the component with the worst heat dissipation capability for server development and testing, thereby improving the quality of server heat dissipation development and testing. The device is also equipped with a switch, and the main controller is connected to the switch via a network cable; The switch is connected to the network ports of each server inside the incubator. Two servers are set up inside the temperature chamber; all components to be compared are installed and tested in pairs at the same location on the servers, and the record information of the component with the highest temperature is output.

2. A method for comparing the heat dissipation capabilities of components in a server, characterized in that, The method is based on the apparatus of claim 1, and the method includes the following steps: Install the components to be compared in the same location on the server and connect the temperature sensors. The main controller sets the temperature of the chamber via signal lines; The main controller controls the server to power on and run the stress test program for the components via a network cable. Temperature sensors acquire the temperature of components in each server and transmit it to the main controller; The main controller obtains the highest temperature of the component at the same location on each server and sets it as the reference temperature, and sets the corresponding component as the reference component; and records the identification code and reference temperature of the reference component; Output the recorded information of the reference components at all locations to be tested on the server; After the temperature sensors acquire the temperature of each component in the server and transmit it to the main controller, the following steps are included: The main controller determines whether the reference temperature exists; If so, compare the temperature of each component at the same location on the server with the reference temperature. If the temperature of a component is higher than the reference temperature, execute the following steps: The main controller obtains the highest temperature of each component at the same location on the server and sets it as the reference temperature, and sets the corresponding component as the reference component; and records the identification code and reference temperature of the reference component. If not, keep the original reference temperature and corresponding reference components unchanged.

3. The method for comparing the heat dissipation capabilities of components in a server according to claim 2, characterized in that, The main controller obtains the highest temperature of components at the same location on each server and sets it as the reference temperature, designating the corresponding component as the reference component; after recording the identification code and reference temperature of the reference component, the following steps are included: Determine if all components in the same location on the server have completed testing; If so, output the record information of the benchmark component in the same location of the server to the test list; If not, reselect the component to be compared at that location and perform the following steps: install the components to be compared in the same location on the server and connect the temperature sensor.

4. The method for comparing the heat dissipation capabilities of components in a server according to claim 3, characterized in that, The step of outputting the record information of the benchmark component in the same location of the server to the test list includes: Determine whether the testing of all components to be compared at all test locations has been completed; If so, the following steps are executed: Output the record information of the reference components at all test locations on the server, that is, output the test list; If not, proceed as follows: Install the components to be compared in the same location on the server and connect the temperature sensors.

Images