Computer monitoring management method and system

By designing the connection between the programmable controller and the processing unit and the storage unit in a computer system, the temperature monitoring logic is integrated, and the complexity and management difficulty problems caused by the unintegration of modules in the prior art are solved, thereby improving system stability and reliability , simplifies the maintenance process.

CN118897777APending Publication Date: 2024-11-05INSPUR SUZHOU INTELLIGENT TECH CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411377335.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-30
Publication Date
2024-11-05

AI Technical Summary

Technical Problem

In the prior art, the overheating management module, temperature warning module and memory thermal management module of the processor are not fully integrated, resulting in an increase in hardware line complexity and increased management difficulty.

Method used

Design a computer monitoring management method and system, and connect it to the processing unit and storage unit through a programmable controller to realize temperature monitoring and management, including generating frequency down signal, turning off power supply and other mechanisms, integrating temperature monitoring logic, and adding new storage units overtemperature monitoring.

Benefits of technology

The logical integration of various temperature monitoring mechanisms is realized, the stability and reliability of the system is improved, the failure rate is reduced, the maintenance work is simplified, and time and cost are saved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118897777A_ABST
    Figure CN118897777A_ABST
Patent Text Reader

Abstract

The invention provides a computer monitoring management method and system, and relates to the technical field of computers.The method comprises the steps that under the condition that the temperature of a monitoring processing unit exceeds a first threshold value, a first control signal is generated and sent to external equipment, so that the external equipment operates at a reduced frequency; under the condition that the temperature of the monitoring processing unit exceeds a second threshold value, a second frequency reduction signal is generated and sent to the processing unit, so that the processing unit actively executes frequency reduction operation; under the condition that the temperature of the processing unit exceeds a third threshold value, an over-temperature signal sent by the processing unit is received, and power supply of the processing unit is closed; when the temperature of the storage unit is monitored to exceed a fourth threshold value, a temperature early warning signal is sent to the processing unit, and a temperature reduction signal sent by the processing unit is received to reduce the power supply of the storage unit, so that the logic of each temperature monitoring mechanism is integrated into a unified temperature monitoring system; and the stability and the reliability of the system are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a computer monitoring management method and system. Background Art

[0002] With the rapid development of information technology, servers, as the core of data processing and storage, are facing unprecedented development opportunities. The popularity of cloud computing, big data, and artificial intelligence has stimulated strong demand for high-performance servers. The rise of edge computing allows servers to perform closer to data sources and speed up response speed.

[0003] In servers, temperature monitoring and management of CPU and memory are crucial to the stable operation of servers. As the core component of modern information technology infrastructure, servers carry a large amount of data processing tasks. Excessive temperature can lead to hardware performance degradation, shortened lifespan, and even failure, which seriously affects service continuity and data security. Therefore, implementing an effective temperature monitoring mechanism can help IT managers detect and solve potential overheating problems in a timely manner, ensuring system reliability and efficiency.

[0004] In the current AMD platform server design, the CPU overheat management module and temperature warning module are usually designed as independent modules, each of which is responsible for different thermal management and power control tasks. However, these modules have not been fully integrated into one, resulting in increased logic complexity in hardware circuits or programmable logic devices, and increased difficulty in management and maintenance. Summary of the invention

[0005] The present invention provides a computer monitoring management method and system to solve the defects in the prior art that the overheating management module, temperature warning module and thermal management module of the processor are not fully integrated into one, resulting in increased logic complexity in hardware circuits or programmable logic devices and increased difficulty in management and maintenance.

[0006] The present invention provides a computer monitoring management method, wherein the computer comprises a processing unit and a storage unit, and the computer monitoring management system comprises a programmable controller, wherein the programmable controller is connected to the processing unit and the storage unit respectively; The method is used for a programmable controller and comprises: When the temperature of the processing unit exceeds a first threshold, generating a first control signal and sending the first control signal to an external device so that the external device can operate at a lower frequency; When the temperature of the processing unit exceeds a second threshold, generating a second frequency reduction signal and sending it to the processing unit, so that the processing unit actively performs a frequency reduction operation; When the temperature of the processing unit exceeds a third threshold, receiving an over-temperature signal sent by the processing unit, and shutting down the power supply of the processing unit; wherein the first threshold, the second threshold, and the third threshold increase in sequence; When the temperature of the storage unit is monitored to exceed a fourth threshold, a temperature warning signal is sent to the processing unit, and a temperature reduction signal sent by the processing unit is received to reduce power supply to the storage unit.

[0007] According to the computer monitoring management method of the processing unit and the storage unit provided by the present invention, the programmable controller includes a management chip, a power supply management chip and a mainboard management chip; the processing unit, the power supply management chip and the mainboard management chip are respectively connected to the management chip; the storage unit is respectively connected to the power supply management chip and the processing unit; The method is used for the management chip, and when the temperature of the monitoring processing unit exceeds a first threshold, generates a first control signal and sends it to an external device, so that the external device reduces the frequency of operation, specifically including: When the temperature of the processing unit monitored by the motherboard management chip exceeds a first threshold, receiving a first control signal sent by the motherboard management chip, generating a first frequency reduction signal and sending it to an external device, so that the external device can operate at a reduced frequency; When the temperature of the processing unit exceeds a second threshold, generating a second frequency reduction signal and sending it to the processing unit so that the processing unit actively performs a frequency reduction operation, specifically comprising: When the temperature of the processing unit monitored by the power management chip exceeds a second threshold, an overheating signal sent by the power management chip is received, and a second frequency reduction signal is generated and sent to the processing unit, so that the processing unit actively performs a frequency reduction operation; When the temperature of the processing unit exceeds a third threshold, receiving an over-temperature signal sent by the processing unit and shutting down the power supply of the processing unit specifically includes: When the temperature of the processing unit exceeds a third threshold, an over-temperature signal sent by the processing unit is received, and a second control signal is generated and sent to the power supply management chip, so that the power supply management chip generates an enable signal to shut down the power supply of the processing unit; When the temperature of the storage unit exceeds a fourth threshold, a temperature warning signal is sent to the processing unit, and a temperature reduction signal sent by the processing unit is received to reduce the power supply of the storage unit, specifically including: When the power management chip monitors that the temperature of the storage unit exceeds a fourth threshold, a temperature warning signal issued by the power management chip is received and sent to the processing unit, so that the processing unit generates a temperature reduction signal to the power management chip to reduce the power supply of the storage unit.

[0008] According to the computer monitoring management method provided by the present invention, the management chip is further connected to a power supply unit; Before generating the first frequency reduction signal and sending it to the processing unit, the method further includes: When the temperature of the power supply unit exceeds a fifth threshold, receiving a power supply temperature alarm signal sent by the power supply unit; Generating a first frequency reduction signal and sending it to an external device specifically includes: The first frequency reduction signal is generated according to the power supply temperature alarm signal and / or the overheating signal and sent to the external device.

[0009] According to the computer monitoring management method of the processing unit and the storage unit provided by the present invention, the management chip includes a first AND gate; Receiving a first control signal sent by the mainboard management chip, generating a first frequency reduction signal and sending it to an external device, specifically includes: receiving the first control signal sent by the mainboard management chip through the first input end of the first AND gate; receiving an interrupt signal sent by the mainboard management chip through the second input end of the first AND gate; generating a first frequency reduction signal according to the first control signal and the interrupt signal through the first AND gate, and sending the first frequency reduction signal to the external device via the output end of the first AND gate.

[0010] According to the computer monitoring management method provided by the present invention, the management chip includes a first AND gate, a second AND gate and a mixer, and the power supply units are two; Receiving a power supply temperature alarm signal sent by the power supply unit specifically includes: The two power supply units are connected respectively through the first input terminal and the second input terminal of the second AND gate to respectively receive the power supply temperature alarm original signals of the two power supply units, and generate a power supply temperature alarm summary signal to be output to the first input terminal of the mixer via the output terminal of the second AND gate; Generating the frequency reduction signal according to the power supply temperature alarm signal and / or the overheat signal and sending it to the external device specifically includes: Receiving a fault alarm signal of the mainboard management chip through the control end of the mixer; Input a default high-level signal through the second input terminal of the mixer, so that the mixer generates a power supply temperature alarm signal according to the power supply temperature alarm summary signal, the fault alarm signal and the high-level signal, and outputs the power supply temperature alarm signal to the second input terminal of the first AND gate through the output terminal of the mixer; Receiving a frequency reduction signal sent by the mainboard management chip BMC through the first input end of the first AND gate; receiving an interrupt signal sent by the mainboard management chip BMC and / or a power supply temperature alarm signal sent by the mixer through the second input end of the first AND gate; A first frequency reduction signal is generated by the first AND gate according to the frequency reduction signal, the interrupt signal and / or the power supply temperature alarm signal, and the first frequency reduction signal is sent to the external device via the output end of the first AND gate.

[0011] According to the computer monitoring management method of the processing unit and the storage unit provided by the present invention, the power supply management chip includes a processing unit power supply management chip, and the management chip includes a buffer and a third AND gate; When the temperature of the processing unit monitored by the power management chip exceeds a second threshold, receiving an overheating signal sent by the power management chip, and generating a second frequency reduction signal to send to the processing unit specifically includes: When the temperature of the processing unit monitored by the processing unit power supply management chip exceeds a second threshold, receiving an overheating signal sent by the processing unit power supply management chip through the buffer; Inputting the overheat signal to the first input terminal of a third AND gate through the buffer, and inputting the first frequency reduction signal to the second input terminal of the third AND gate through the first AND gate; The second frequency reduction signal is generated according to the overheat signal and the first frequency reduction signal through the third AND gate, and the second frequency reduction signal is sent to the corresponding processing unit.

[0012] According to the computer monitoring management method provided by the present invention, the processing unit power supply management chip is also used to perform overcurrent protection on the processing unit; The method also includes: when the current of the processing unit monitored by the processing unit power supply management chip is greater than a first current threshold, the processing unit power supply management chip directly sends a first overcurrent protection signal to the processing unit to stop the processing unit and cut off the power supply.

[0013] According to the computer monitoring management method provided by the present invention, the management chip includes a fourth AND gate and a trigger; the trigger is connected to the mainboard management chip, and the fourth AND gate is respectively connected to the trigger and the processing unit; When the temperature of the processing unit exceeds a third threshold, an over-temperature signal sent by the processing unit CPU is received, and a second control signal is generated and sent to the power supply management chip, specifically including: When the temperature of the processing unit exceeds a third threshold, receiving an over-temperature signal sent by the processing unit through the first input terminal of the fourth AND gate; receiving a power-on signal sent by the processing unit through the second input terminal of the fourth AND gate; Generate a trigger control signal according to the over-temperature signal and the power-on signal through the fourth AND gate, and send the signal to the trigger; A corresponding trigger signal is generated by the trigger according to the trigger control signal, and the trigger signal is sent to the power supply management chip.

[0014] According to the computer monitoring management method provided by the present invention, the management chip includes a buffer, and the buffer is connected to the mainboard management chip; When the temperature of the processing unit exceeds a third threshold, the method further includes: The over-temperature signal sent by the processing unit is received through the buffer, so that the mainboard management chip reads the over-temperature signal and issues a warning.

[0015] According to the computer monitoring management method provided by the present invention, the power supply management chip includes a storage unit power supply management chip, the management chip includes a buffer, and the storage unit power supply management chip is connected to the buffer; When the temperature of the storage unit monitored by the power management chip exceeds a fourth threshold, receiving a temperature warning signal issued by the power management chip and sending the signal to the processing unit, so that the processing unit generates a temperature reduction signal to the power management chip, specifically comprising: When the storage unit power supply management chip monitors that the temperature of the storage unit exceeds a fourth threshold, the buffer receives a temperature warning signal issued by the storage unit power supply management chip and sends it to the processing unit, so that the processing unit generates a temperature reduction signal to the storage unit power supply management chip.

[0016] According to the computer monitoring management method provided by the present invention, the storage unit power supply management chip is also used for overcurrent protection of the storage unit; The method also includes: when the current of the storage unit monitored by the storage unit power supply management chip is greater than a second current threshold, directly sending a second overcurrent protection signal to the processing unit through the storage unit power supply management chip to cause the processing unit to disconnect the power supply of the storage unit.

[0017] The present invention provides a computer monitoring and management system, wherein the computer comprises a processing unit and a storage unit, and comprises a programmable controller, wherein the programmable controller is connected to the processing unit and the storage unit respectively and is configured as follows: In response to the temperature of the processing unit exceeding a first threshold, generating a first control signal and sending the first control signal to an external device, so that the external device operates at a reduced frequency; In response to the temperature of the processing unit exceeding a second threshold, generating a second frequency reduction signal and sending it to the processing unit, so that the processing unit actively performs a frequency reduction operation; In response to an over-temperature signal sent by the processing unit, turning off the power supply of the processing unit; wherein the over-temperature signal sent by the processing unit is generated when the temperature of the processing unit exceeds a third threshold, and the first threshold, the second threshold and the third threshold increase in sequence; In response to monitoring that the temperature of the storage unit exceeds a fourth threshold, a temperature warning signal is sent to the processing unit, and a temperature reduction signal sent by the processing unit is received to reduce the power supply of the storage unit.

[0018] Optionally, the programmable controller includes a management chip, a power supply management chip and a mainboard management chip; the processing unit, the power supply management chip and the mainboard management chip are respectively connected to the management chip; the storage unit is respectively connected to the power supply management chip and the processing unit; The temperature of the processing unit is monitored by the mainboard management chip, and in response to the temperature of the processing unit exceeding a first threshold, a first control signal sent by the mainboard management chip is received by the management chip, a first frequency reduction signal is generated and sent to an external device, so that the external device can operate at a reduced frequency; The power supply management chip monitors the temperature of the processing unit, and in response to the temperature of the processing unit exceeding a second threshold, the management chip receives an overheating signal sent by the power supply management chip, generates a second frequency reduction signal and sends it to the processing unit, so that the processing unit actively performs a frequency reduction operation; In response to receiving the over-temperature signal sent by the processing unit, generating a second control signal through the management chip and sending it to the power management chip, so that the power management chip generates an enable signal to turn off the power supply of the processing unit; The temperature of the storage unit is monitored by the power supply management chip. In response to exceeding a fourth threshold, a temperature warning signal issued by the power supply management chip is received by the management chip and sent to the processing unit, so that the processing unit generates a temperature reduction signal to the power supply management chip to reduce power supply to the storage unit.

[0019] According to the computer monitoring and management system provided by the present invention, the management chip is further connected to a power supply unit; In response to the temperature of the power supply unit exceeding the fifth threshold, the management chip receives a power temperature alarm signal sent by the power supply unit, and generates the first frequency reduction signal based on the power temperature alarm signal and / or the overheating signal and sends it to the external device.

[0020] According to the computer monitoring and management system of the processing unit and the storage unit provided by the present invention, the management chip includes a first AND gate; A first control signal sent by the mainboard management chip is received through the first input end of the first AND gate, and an interrupt signal sent by the mainboard management chip is received through the second input end of the first AND gate; a first frequency reduction signal is generated according to the first control signal and the interrupt signal through the first AND gate, and the first frequency reduction signal is sent to the external device via the output end of the first AND gate, so that the external device can operate at a reduced frequency.

[0021] According to the computer monitoring and management system provided by the present invention, the management chip includes a first AND gate, a second AND gate and a mixer, and the number of the power supply units is two; The two power supply units are connected respectively through the first input terminal and the second input terminal of the second AND gate to respectively receive the power supply temperature alarm original signals of the two power supply units, and generate a power supply temperature alarm summary signal to be output to the first input terminal of the mixer via the output terminal of the second AND gate; Receiving a fault alarm signal of the mainboard management chip through the control end of the mixer; Input a default high-level signal through the second input terminal of the mixer, so that the mixer generates a power supply temperature alarm signal according to the power supply temperature alarm summary signal, the fault alarm signal and the high-level signal, and outputs the power supply temperature alarm signal to the second input terminal of the first AND gate through the output terminal of the mixer; Receiving a frequency reduction signal sent by the mainboard management chip through a first input end of the first AND gate, and receiving an interrupt signal sent by the mainboard management chip and / or a power supply temperature alarm signal sent by the mixer through a second input end of the first AND gate; A first frequency reduction signal is generated by the first AND gate according to the frequency reduction signal, the interrupt signal and / or the power supply temperature alarm signal, and the first frequency reduction signal is sent to the external device via the output end of the first AND gate.

[0022] According to the computer monitoring and management system provided by the present invention, the power supply management chip includes a processing unit power supply management chip, and the management chip includes a buffer and a third AND gate; monitoring the temperature of the processing unit through the processing unit power management chip, and in response to the temperature of the processing unit exceeding a second threshold, receiving an overheating signal sent by the processing unit power management chip through the buffer; Inputting the overheat signal to the first input terminal of a third AND gate through the buffer, and inputting the first frequency reduction signal to the second input terminal of the third AND gate through the first AND gate; The second frequency reduction signal is generated according to the overheat signal and the first frequency reduction signal through the third AND gate, and the second frequency reduction signal is sent to the corresponding processing unit.

[0023] According to the computer monitoring and management system provided by the present invention, the management chip includes a fourth AND gate and a trigger; the trigger is connected to the mainboard management chip, and the fourth AND gate is connected to the trigger and the processing unit respectively; Receiving the over-temperature signal through the first input terminal of the fourth AND gate, receiving the power-on signal sent by the processing unit through the second input terminal of the fourth AND gate, generating a first frequency reduction signal according to the over-temperature signal and the power-on signal through the fourth AND gate, and sending the first frequency reduction signal to the trigger; A corresponding trigger signal is generated according to the first frequency reduction signal by the trigger, and the trigger signal is sent to the power supply management chip.

[0024] According to the computer monitoring and management system provided by the present invention, the power supply management chip includes a storage unit power supply management chip, the management chip includes a buffer, and the storage unit power supply management chip is connected to the buffer; The power supply management chip of the storage unit monitors the temperature of the storage unit, and in response to the temperature of the storage unit exceeding a fourth threshold, the power supply management chip generates a temperature warning signal; The temperature warning signal is received by the buffer and sent to the processing unit, so that the processing unit generates a temperature reduction signal to the storage unit power supply management chip.

[0025] The computer monitoring management method and system provided by the present invention, when the temperature of the monitoring processing unit exceeds the first threshold, generates a first control signal and sends it to an external device, so that the external device reduces the frequency of operation, thereby realizing the primary temperature protection mechanism of the processing unit; when the temperature of the monitoring processing unit exceeds the second threshold, generates a second frequency reduction signal and sends it to the processing unit, so that the processing unit actively performs the frequency reduction operation, thereby realizing the temperature early warning protection mechanism of the processing unit; when the temperature of the processing unit exceeds the third threshold, receives the over-temperature signal sent by the processing unit, shuts down the power supply of the processing unit, thereby realizing the over-temperature protection mechanism of the processing unit; when the temperature of the monitoring storage unit exceeds the fourth threshold, sends a temperature early warning signal to the processing unit, and receives the temperature reduction signal sent by the processing unit, thereby reducing the power supply of the storage unit, thereby realizing the over-temperature protection mechanism of the storage unit; the scheme of the present invention thus realizes the integration of the logic of each temperature monitoring mechanism into a unified temperature monitoring system, and adds over-temperature monitoring of the storage unit, thereby improving the stability and reliability of the system, reducing the failure rate, simplifying the maintenance work, and saving time and cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0027] Figure 1 It is a schematic diagram of the framework of the computer monitoring and management system provided by the present invention.

[0028] Figure 2 It is a flow chart of the computer monitoring management method provided by the present invention.

[0029] Figure 3 It is a detailed circuit diagram of the computer monitoring and management system provided by the present invention.

[0030] Figure 4 It is a schematic diagram of the hardware design topology circuit corresponding to the PROCHOT and THROTTLE mechanisms provided by the present invention.

[0031] Figure 5 It is a schematic diagram of the hardware design topology circuit of the THERMTRIP and APML mechanisms provided by the present invention.

[0032] Figure 6 It is a schematic diagram of the hardware design topology circuit corresponding to the MEMHOT mechanism provided by the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] First, the terms involved in the embodiments of the present invention are schematically explained.

[0035] CPLD (Complex Programmable Logic Device): Complex programmable logic device.

[0036] MUX: Mixer, a chip used to implement signal link switching.

[0037] CPU VR Controllers: Processor power management chip.

[0038] MEM VR Controllers: memory power management chip.

[0039] BMC (Board Management Chip): mainboard management chip.

[0040] Filter: A filter module in CPLD.

[0041] PCIE (Peripheral Component Interconnect Express): High-speed serial computer expansion bus.

[0042] THERMTRIP: A temperature emergency stop signal that is triggered when the processor temperature reaches a critical temperature to prevent hardware damage.

[0043] PROCHOT: A temperature warning signal indicating that the processor temperature is close to but has not reached the THERMTRIP trigger point, allowing the system to take preventive measures in advance.

[0044] THROTTLE: Automatically reduces processor performance to reduce heat and ensure stable server operation.

[0045] APML (Advanced Platform Management Link): monitors high memory temperatures and avoids memory failures by adding cooling measures. APML provides an interface for remotely monitoring and managing the system's power usage and cooling status, enabling more sophisticated power management.

[0046] MEMHOT: Memory temperature management mechanism, used to provide early warning signals for memory temperature.

[0047] In the design of the server, THERMTRIP, PROCHOT, THROTTLE, MEMHOT and APML management mechanisms together constitute the CPU and memory temperature monitoring and management system in the server, which is a crucial temperature management and performance control mechanism in the server.

[0048] Through the solution of the embodiment of the invention, the system after integrating the THERMTRIP, PROCHOT, THROTTLE, APML and MEMHOT mechanisms is more efficient. The originally scattered mechanisms need to be managed separately, but the solution of the embodiment of the invention is centrally controlled through a unified logical structure, which reduces unnecessary line complexity and management difficulty. The system responds faster, the resource utilization rate is higher, and the maintenance cost is also reduced.

[0049] First, a computer monitoring and management system according to an embodiment of the present invention is schematically described. Figure 1 The computer monitoring and management system includes a programmable controller, specifically including: a management chip, a power supply management chip and a mainboard management chip. The computer includes a processing unit and a storage unit. The processing unit, the power supply management chip and the mainboard management chip are respectively connected to the management chip.

[0050] The storage unit is connected to the power management chip and the processing unit respectively. When the power management chip detects that the temperature of the storage unit exceeds a threshold, it sends a temperature warning signal to the processing unit, so that the processing unit generates a temperature reduction signal to the power management chip to reduce the power supply of the storage unit.

[0051] The external device is connected to the management chip and is used for receiving the first frequency reduction signal sent by the management chip.

[0052] In this embodiment, specifically, the management chip may be a programmable logic device, the processing unit may be a processor CPU, and the storage unit may be a memory.

[0053] The method of the embodiment of the present invention is shown in Figure 2 ,include: 201. When the temperature of the processing unit is monitored to be above a first threshold, generate a first control signal and send it to an external device, so that the external device can operate at a reduced frequency.

[0054] Specifically, when the management chip monitors the temperature of the processing unit through the mainboard management chip and the temperature exceeds a first threshold, the management chip receives a first control signal sent by the mainboard management chip, generates a first frequency reduction signal and sends it to the external device, so that the external device reduces the frequency of operation.

[0055] Specifically, the computer monitoring management system of the embodiment of the present invention is described in detail. Figure 3 The mainboard management chip BMC generates a first control signal BMC_FORCE_THROTTLE_N and sends it to the programmable logic device CPLD. The programmable logic device CPLD generates a first frequency reduction signal FM_SYS_THROTTLE_N and sends it to the external device.

[0056] The external device in this embodiment may be a PCIE device, and the PCIE device is connected to the programmable logic device CPLD via a serial bus to receive the first frequency reduction signal sent by the programmable logic device CPLD.

[0057] The processor and the mainboard management chip are connected via a bus so that the mainboard management chip can obtain the temperature of the processor.

[0058] When external devices run at a lower frequency, the burden on the processor is reduced, thus alleviating processor problems to a certain extent.

[0059] 202. When the temperature of the processing unit is monitored to be greater than a second threshold, generate a second frequency reduction signal and send it to the processing unit, so that the processing unit actively performs a frequency reduction operation.

[0060] Specifically, when the management chip monitors the temperature of the processing unit through the power management chip and the temperature exceeds the second threshold, the management chip receives an overheating signal sent by the power management chip, generates a second frequency reduction signal and sends it to the processing unit, so that the processing unit actively performs a frequency reduction operation.

[0061] When the processor temperature continues to rise to exceed the second threshold, the power management chip will monitor that the processor temperature exceeds the second threshold (for example, 95 degrees), generate an overheating signal VR_ALERT_N and send it to the programmable logic device CPLD. The programmable logic device CPLD generates a second frequency reduction signal PROCHOT_L to the processor, so that the processor actively performs a frequency reduction operation to reduce heat generation.

[0062] 203. When the temperature of the processing unit exceeds a third threshold, receive an over-temperature signal sent by the processing unit, and turn off power supply to the processing unit.

[0063] Specifically, when the temperature of the processing unit exceeds a third threshold, the management chip receives an over-temperature signal sent by the processing unit, and generates a second control signal and sends it to the power management chip, so that the power management chip generates an enable signal to shut down the power supply of the processing unit.

[0064] If the processor temperature continues to rise to exceed the third threshold (eg, 115 degrees), the programmable logic device CPLD receives the over-temperature signal THERMTRIP_L sent by the processor, generates a second control signal and sends it to the power management chip, so that the power management chip generates an enable signal VR_OCP_N to shut down the power supply of the processor.

[0065] Among them, the first threshold, the second threshold and the third threshold increase in sequence.

[0066] In this embodiment, in one usage scenario, the first threshold value may be set to 80 degrees, the second threshold value may be set to 95 degrees, and the third threshold value may be set to 115 degrees.

[0067] 204. When the temperature of the storage unit monitored exceeds a fourth threshold, send a temperature warning signal to the processing unit, and receive a temperature reduction signal sent by the processing unit to reduce power supply to the storage unit.

[0068] Specifically, when the power management chip monitors that the temperature of the storage unit exceeds a fourth threshold, a temperature warning signal issued by the power management chip is received and sent to the processing unit, so that the processing unit generates a temperature reduction signal to the power management chip to reduce the power supply of the storage unit.

[0069] The temperature monitoring of the memory and the temperature monitoring of the processor are performed in parallel.

[0070] The fourth threshold can be adjusted according to actual needs, for example, the fourth threshold is set to 80 degrees.

[0071] In the embodiment of the present invention, specifically, the management chip may be a programmable logic device, the processing unit may be a processor CPU, and the storage unit may be a memory.

[0072] The computer monitoring and management method provided by the embodiment of the present invention generates a first control signal and sends it to an external device when the temperature of the processing unit is monitored to exceed a first threshold value, so that the external device reduces the frequency of operation, thereby realizing the primary temperature protection mechanism of the processing unit; generates a second frequency reduction signal and sends it to the processing unit when the temperature of the processing unit is monitored to exceed a second threshold value, so that the processing unit actively performs a frequency reduction operation, thereby realizing the temperature warning protection mechanism of the processing unit; receives an over-temperature signal sent by the processing unit when the temperature of the processing unit exceeds a third threshold value, shuts off the power supply of the processing unit, thereby realizing the over-temperature protection mechanism of the processing unit; sends a temperature warning signal to the processing unit when the temperature of the storage unit is monitored to exceed a fourth threshold value, receives a temperature reduction signal sent by the processing unit, and receives a temperature reduction signal sent by the processing unit, thereby reducing the power supply of the storage unit, thereby realizing the over-temperature protection mechanism of the storage unit, thereby realizing the integration of the logic of each temperature monitoring mechanism into a unified temperature monitoring system, and adding over-temperature monitoring of the memory, thereby improving the stability and reliability of the system, reducing the failure rate, simplifying maintenance work, and saving time and cost.

[0073] Furthermore, the management chip (programmable logic device CPLD) is further connected to the power supply unit, and is used to consider the situation that the temperature of the power supply unit exceeds the fifth threshold value in the process of generating the first frequency reduction signal. Correspondingly, before generating the first frequency reduction signal and sending it to the processor CPU, the method also includes: when the temperature of the power supply unit PSU exceeds the fifth threshold value, receiving a power supply temperature alarm signal sent by the power supply unit PSU; In step 201, generating a first frequency reduction signal and sending it to an external device specifically includes: generating a first frequency reduction signal and sending it to an external device according to a power temperature alarm signal and / or an overheating signal.

[0074] In addition to processor temperature monitoring, the embodiment of the present invention also includes temperature monitoring of a power supply unit (PSU) to achieve more comprehensive system thermal management.

[0075] The following embodiments respectively schematically illustrate the primary temperature protection mechanism of the processor, the temperature warning protection mechanism of the processor, the over-temperature protection mechanism of the processor, and the over-temperature protection mechanism of the memory.

[0076] The processor's primary temperature protection mechanism and the processor's temperature warning protection mechanism use the PROCHOT and THROTTLE mechanisms.

[0077] The processor's over-temperature protection mechanism uses THERMTRIP and APML mechanisms.

[0078] The memory's over-temperature protection mechanism uses the MEMHOT mechanism.

[0079] First, the primary temperature protection mechanism of the processor is schematically explained. Figure 4 , Figure 4 The red circuit shown is the hardware design topology corresponding to the PROCHOT and THROTTLE mechanisms in this embodiment. Specifically, on the AMD platform, when the CPU temperature exceeds 95 degrees but has not reached the 115-degree threshold set by THERMTRIP, the PROCHOT signal will be triggered, reminding the system to take cooling measures in advance. The THROTTLE mechanism will automatically reduce the performance of the processor when necessary to reduce heat generation and ensure stable operation of the server. The design of the PROCHOT module acts on the CPU through external triggering, thereby achieving throttling of the system and reducing the CPU temperature.

[0080] Optionally, the programmable logic device includes a first AND gate D1. In step 201, the first control signal sent by the motherboard management chip is received, and a first frequency reduction signal is generated and sent to an external device, specifically including: receiving the first control signal sent by the motherboard management chip BMC through the first input end of the first AND gate D1; receiving an interrupt signal sent by the motherboard management chip BMC through the second input end of the first AND gate D1; generating a first frequency reduction signal according to the first control signal and the interrupt signal through the first AND gate D1, and sending the first frequency reduction signal to the external device via the output end of the first AND gate D1.

[0081] Optionally, the programmable logic device includes a first AND gate D1, a second AND gate D2 and a mixer MUX, and there are two power supply units PSU.

[0082] Receiving the power supply temperature alarm signal sent by the power supply unit PSU, specifically comprising: connecting the two power supply units PSU through the first input terminal and the second input terminal of the second AND gate D2 respectively, so as to receive the power supply temperature alarm original signals of the two power supply units PSU respectively, and generating a power supply temperature alarm summary signal and outputting it to the first input terminal of the mixer MUX via the output terminal of the second AND gate D2; Generating the frequency reduction signal according to the power supply temperature alarm signal and / or the overheat signal and sending it to the external device specifically includes: The fault alarm signal FM_PMBUS_ALERT_BMC_N of the mainboard management chip BMC is received through the control end of the mixer MUX; if the fault alarm signal is 0, the mixer can output the output signal (low level) of the second AND gate D2 to the first AND gate D1; if the fault alarm signal is 1, the mixer outputs the default high level signal to the first AND gate D1.

[0083] The default high level signal 1b1 is input through the second input terminal of the mixer MUX, so that the mixer MUX generates a power temperature alarm signal according to the power temperature alarm summary signal, the fault alarm signal and the high level signal, and outputs it to the second input terminal of the first AND gate D1 through the output terminal of the mixer.

[0084] The frequency reduction signal sent by the mainboard management chip BMC is received through the first input end of the first AND gate D1; the interrupt signal sent by the mainboard management chip BMC and / or the power supply temperature alarm signal sent by the mixer MUX is received through the second input end of the first AND gate D1.

[0085] The first AND gate D1 generates a first frequency reduction signal FM_SYS_THROTTLE_N according to the frequency reduction signal, the interrupt signal and / or the power temperature alarm signal, and sends the first frequency reduction signal FM_SYS_THROTTLE_N to the external device via the output end of the first AND gate D1.

[0086] like Figure 4 As shown, the BMC continuously monitors the temperature of the CPU and the system. Once the CPU temperature is found to be too high, the BMC will send a first control signal BMC_FORCE_THROTTLE_N to the CPLD. The signal first passes through the Filter filtering module and then serves as the input of the first AND gate D1. At the same time, the original power temperature alarm signal PMBUS ALERT on the server's power supply unit PSU (Power Supply Unit) will also be triggered and sent to the second AND gate D2 when the PSU temperature is detected to be too high. The signal after passing through the second AND gate D2 will enter the mixer MUX, and the fault alarm signal FM_PMBUS_ALERT_BMC_N issued by the BMC will serve as the SEL (Select) signal of the MUX. After a series of logical processing by the first AND gate D1, the second AND gate D2 and the mixer MUX, the generated first frequency reduction signal is sent to the PCIE device, causing the PCIE device to reduce the frequency, thereby reducing the computing burden of the CPU.

[0087] Next, the temperature early warning protection mechanism of this embodiment is schematically described below.

[0088] Continue to see Figure 4 , the power supply management chip includes a processor power supply management chip, and the programmable logic device includes a buffer and a third AND gate; When the temperature of the processor is monitored by the power management chip and exceeds a second threshold, an overheating signal sent by the power management chip CPU VR is received, and a second frequency reduction signal is generated and sent to the processor CPU, specifically including: In a case where the temperature of the processor monitored by the processor power management chip exceeds a second threshold, receiving an overheating signal sent by the processor power management chip through the buffer; Inputting the overheat signal to the first input terminal of a third AND gate through the buffer, and inputting the first frequency reduction signal to the second input terminal of the third AND gate through the first AND gate; The second frequency reduction signal is generated according to the overheat signal and the first frequency reduction signal through the third AND gate, and the second frequency reduction signal is sent to the corresponding processor.

[0089] When the CPU temperature gradually rises and triggers PROCHOT, the temperature of the processor power management chip CPU VR Controllers also rises. When the processor power management chip sends out an overheating signal VR_ALERT_N, the signal is not only stored in the register Register, but also input into the third AND gate. After the logic operation of the third AND gate, the second frequency reduction signal is generated and output to the CPU's PROCHOT_L Port, prompting the CPU to actively perform frequency reduction operations.

[0090] In addition, the processor power management chip is also used to protect the processor from overcurrent; The method also includes: when the current of the processor monitored by the processor power management chip is greater than a first current threshold, the processor power management chip directly sends a first overcurrent protection signal to the processor to stop the processor and cut off the power supply.

[0091] If the processor power management chip CPU VR triggers overcurrent protection, the VR_OCP_N signal will be sent directly to the CPU's PCC port via the CPLD. At this time, in order to prevent the processor power management chip CPU VR from being damaged, the processor power management chip CPU VR will immediately cut off the power supply, and the processor CPU will stop working. The information of this process will be recorded in the register in the CPLD. The BMC can read the contents of the register through the I2C bus and record it in the log, which is convenient for maintenance personnel to quickly locate the cause of power failure and perform troubleshooting.

[0092] Again, the over-temperature protection mechanism of this embodiment is schematically described. Figure 5 , Figure 5 The red circuit portion shows the hardware design topology of the THERMTRIP and APML mechanisms of this embodiment.

[0093] Optionally, the programmable logic device includes a fourth AND gate D4 and a trigger Trigger; the trigger Trigger is connected to the mainboard management chip BMC, and the fourth AND gate D4 is connected to the trigger Trigger and the processor CPU respectively.

[0094] When the temperature of the processor exceeds the third threshold, an over-temperature signal sent by the processor CPU is received, and a second control signal is generated and sent to the power management chip, specifically including: when the temperature of the processor exceeds the third threshold (115 degrees), an over-temperature signal THERMTRIP_L sent by the processor is received through the first input end of the fourth AND gate D4; a power-on signal PWROK sent by the processor is received through the second input end of the fourth AND gate D4; a trigger control signal is generated through the fourth AND gate D4 according to the over-temperature signal THERMTRIP_L and the power-on signal PWROK, and sent to the trigger Trigger; a corresponding trigger signal is generated through the trigger Trigger according to the trigger control signal, and the trigger signal is sent to the power management chip CPU VR.

[0095] Specifically, the programmable logic device CPLD includes a buffer Register, and the buffer Register is connected to the mainboard management chip BMC.

[0096] When the temperature of the processor exceeds the third threshold, the method further includes: receiving an over-temperature signal THERMTRIP_L sent by the processor through a buffer Register, so that the mainboard management chip BMC reads the over-temperature signal THERMTRIP_L and issues a warning.

[0097] THERMTRIP is a thermal protection mechanism for servers. When the CPU temperature reaches the preset maximum threshold (115 degrees on the AMD platform), the system will automatically shut down to avoid hardware damage due to overheating. Taking a dual-core server as an example, CPU0 and CPU 1 are each equipped with a THERMTRIP_L Port. When the CPU temperature reaches 115 degrees, the THERMTRIP signal will be pulled to a low level. In this design, the THERMTRIP signals of CPU0 and CPU 1 are both connected to the CPLD on the motherboard. Two AND gate triggers are designed inside the CPLD. The overtemperature signals THERMTRIP of CPU0 and CPU 1 are connected to the AND gates respectively, and the other input of the AND gates is the power-on signal PWROK of CPU0 and CPU 1. Only when the CPU is turned on (PWROK signal triggered) and the overtemperature signal THERMTRIP is also triggered, the AND gate will generate an output, which means that the CPU has reached the threshold temperature and needs to be shut down immediately. In addition, when only one CPU is powered on in a dual-core server, the logic of triggering the over-temperature signal THERMTRIP will not be affected, and this design can also prevent false triggering of the over-temperature signal THERMTRIP.

[0098] When any CPU reaches 115 degrees Celsius, the over-temperature signal THERMTRIP of the CPU is triggered, and when the system is in the power-on state, the AND gate will send a signal to the next-level OR gate. Regardless of which CPU triggers the THERMTRIP mechanism, the OR gate will pass this signal to the trigger Power Off Trigger in the CPLD, which is responsible for starting the power-off shutdown sequence of the CPLD. During the power-off shutdown process, the trigger Power Off Trigger will pull down the enable signal Enable of the processor power management chip CPU VR one by one in a predetermined order, so that the processor power management chip CPU VRControllers will shut down the power supply of the CPU in sequence and complete the system shutdown. In addition, if Figure 5 As shown in the figure, the THERMTRIP signals of CPU0 and CPU1 are also directly transmitted to the register in the CPLD. The BMC can read the data of the register in the CPLD through the I2C bus. If the system shuts down due to the CPU overheating triggering the THERMTRIP protection mechanism, the motherboard management chip BMC will accurately record this event and issue a warning. In this way, maintenance personnel can find the specific reason for the abnormal shutdown in the BMC log - that is, the CPU triggered the THERMTRIP overheating protection mechanism.

[0099] Before the CPU temperature reaches 115 degrees Celsius, the motherboard management chip BMC can track the temperature changes of the CPU in real time through the APML link. Figure 5As shown in the figure, the motherboard management chip BMC is connected to CPU0 using the APML bus. APML follows the I3C bus protocol and is backward compatible with the I2C standard. With the help of the APML link, the motherboard management chip BMC can obtain the temperature reading of the CPU at any time. When the system encounters an abnormal situation and triggers an automatic shutdown, the maintenance personnel can view the CPU temperature gradually rising within a specific time range in the log of the motherboard management chip BMC until the THERMTRIP thermal protection mechanism is triggered, and then the shutdown operation is executed. Such real-time monitoring and recording functions help diagnose the cause of the problem, promptly discover and solve potential hardware failures, and ensure the stable operation of the system.

[0100] Finally, the memory over-temperature protection mechanism of the embodiment of the present invention is schematically described.

[0101] Alternatively, see Figure 6 , Figure 6 The schematic diagram of the hardware design topology circuit corresponding to the MEMHOT mechanism provided in this embodiment is shown. The power supply management chip includes a memory power supply management chip, the programmable logic device includes a buffer, and the memory power supply management chip is connected to the buffer.

[0102] When the temperature of the memory monitored by the memory power management chip exceeds the fourth threshold, the temperature warning signal sent by the power management chip is received by the buffer and sent to the processor, so that the processor generates a temperature reduction signal to the memory power management chip.

[0103] In addition, the memory power management chip is also used for overcurrent protection of the memory. The method also includes: when the current of the memory monitored by the memory power management chip is greater than the second current threshold, the memory power management chip directly sends a second overcurrent protection signal to the processor, so that the processor disconnects the power supply of the memory.

[0104] In an embodiment of the present invention, the MEMHOT mechanism is used to indicate that the temperature of the memory module is too high. During the operation of the server, the memory stick will also generate heat, and overheating may affect performance or cause system instability. In AMD platform servers, although the memory temperature can be monitored through the BMC, it lacks logical processing like PROCHOT and THROTTLE, and it is impossible to determine when the memory should be throttled or when the ALERT alarm should be triggered to disconnect the power supply of a certain memory. Therefore, the solution of this embodiment introduces a memory power supply management chip MEM VR (Memory Voltage Regulator) Controller, which is intended to more effectively monitor and manage the temperature of the memory.

[0105] like Figure 6As shown, when the temperature of the memory power supply management chip MEM VR exceeds the preset threshold, VR will issue a temperature warning signal MEM_VR_ALERT_N. The temperature warning signal MEM_VR_ALERT_N is first stored in the register Register, and then triggers the AND gate, and the output of the AND gate is sent to the interface MEMHOT_IN_L Port of the two processor CPUs respectively. After receiving the signal, the processor CPU sends a temperature reduction signal MEMHOT_N to the corresponding memory power supply management chip MEM VR through the MEMHOT_OUT_L port, thereby reducing the power supply of the memory, reducing the memory calculation amount, and reducing the memory temperature. In addition, if the memory power supply management chip MEM VR triggers the overcurrent protection, the memory power supply management chip MEM VR will issue a second overcurrent protection signal MEM_VR_OCP_L. This signal will be sent to the programmable logic device CPLD and the processor CPU respectively. The register Register in the programmable logic device CPLD will record this information, and the CPU will control the disconnection of the power supply of the corresponding memory to prevent memory damage. The mainboard management chip BMC reads the information of the register in the programmable logic device CPLD through the I2C bus, and records the memory temperature alarm and the memory power management chip MEM VR overcurrent alarm in the log, which is convenient for maintenance personnel to find the cause of the fault.

[0106] Through the design of this solution, the system after integrating the THERMTRIP, PROCHOT, THROTTLE, APML and MEMHOT mechanisms is more efficient. The originally scattered modules need to be managed separately, but now they are centrally controlled through a unified logical structure, reducing unnecessary line complexity and reducing management difficulty. The system responds faster, resource utilization is higher, and maintenance costs are reduced. Secondly, the integrated system has higher flexibility. The new design can flexibly adjust parameters according to actual needs, adapt to different working environments and load conditions, and provide better thermal management strategies. For example, when the CPU temperature approaches the critical value, the system can take preventive measures in advance, such as reducing processor performance, instead of waiting until the limit is reached to trigger an emergency shutdown. Thirdly, the integrated system provides comprehensive temperature monitoring.

[0107] In addition to traditional CPU temperature monitoring, the newly added memory power management chip MEM VR fills the gap in memory temperature monitoring. In this way, the system can monitor the memory temperature in real time to ensure the stability and reliability of the memory. When the memory temperature is too high, the system can take timely measures, such as reducing the memory frequency, to avoid data errors or system crashes caused by overheating. Finally, the integrated system provides detailed fault records. The BMC can read the Register data in the CPLD through the I2C bus and record the reasons for each abnormal shutdown, whether it is CPU or memory overheating, which can be clearly reflected in the log. This helps maintenance personnel to quickly locate problems and improve the speed and accuracy of troubleshooting. In summary, the integrated system integrates multiple thermal management modules, realizes unified temperature monitoring and management, improves the stability and availability of the system, reduces the failure rate, and simplifies maintenance work, saving time and cost.

[0108] The computer monitoring management system provided by an embodiment of the present invention is described below. The computer monitoring management system described below and the computer monitoring management method of the processing unit and the storage unit described above can be referenced to each other.

[0109] An embodiment of the present invention provides a computer monitoring management system, wherein the computer includes a processing unit and a storage unit, and includes a programmable controller, wherein the programmable controller is connected to the processing unit and the storage unit respectively and is configured as follows: In response to the temperature of the processing unit exceeding a first threshold, generating a first control signal and sending the first control signal to an external device, so that the external device operates at a reduced frequency; In response to the temperature of the processing unit exceeding a second threshold, generating a second frequency reduction signal and sending it to the processing unit, so that the processing unit actively performs a frequency reduction operation; In response to an over-temperature signal sent by the processing unit, turning off the power supply of the processing unit; wherein the over-temperature signal sent by the processing unit is generated when the temperature of the processing unit exceeds a third threshold, and the first threshold, the second threshold and the third threshold increase in sequence; In response to monitoring that the temperature of the storage unit exceeds a fourth threshold, a temperature warning signal is sent to the processing unit, and a temperature reduction signal sent by the processing unit is received to reduce the power supply of the storage unit.

[0110] Specifically, see Figure 3, including a programmable logic device CPLD, a power management chip (including a processor power management chip CPU VR Controller and a memory management chip MEM VR Controller) and a motherboard management chip BMC. The processor, the power management chip and the motherboard management chip are respectively connected to the programmable logic device CPLD; the memory MEM is respectively connected to the power management chip and the processor. Specifically, the memory MEM is connected to the memory management chip MEM VR Controller, and the processor is connected to the processor power management chip CPU.

[0111] The temperature of the processing unit is monitored by the mainboard management chip, and in response to the temperature of the processing unit exceeding a first threshold, a first control signal sent by the mainboard management chip is received by the management chip, a first frequency reduction signal is generated and sent to an external device, so that the external device can operate at a reduced frequency; The power supply management chip monitors the temperature of the processing unit, and in response to the temperature of the processing unit exceeding a second threshold, the management chip receives an overheating signal sent by the power supply management chip, generates a second frequency reduction signal and sends it to the processing unit, so that the processing unit actively performs a frequency reduction operation; In response to receiving the over-temperature signal sent by the processing unit, generating a second control signal through the management chip and sending it to the power management chip, so that the power management chip generates an enable signal to turn off the power supply of the processing unit; The temperature of the storage unit is monitored by the power supply management chip. In response to exceeding a fourth threshold, a temperature warning signal issued by the power supply management chip is received by the management chip and sent to the processing unit, so that the processing unit generates a temperature reduction signal to the power supply management chip to reduce power supply to the storage unit.

[0112] Optionally, the programmable logic device is further connected to a power supply unit; In response to the temperature of the power supply unit exceeding the fifth threshold, the management chip receives a power temperature alarm signal sent by the power supply unit, and generates the first frequency reduction signal based on the power temperature alarm signal and / or the overheating signal and sends it to the external device.

[0113] Optionally, the programmable logic device includes a first AND gate D1; A first control signal sent by the mainboard management chip is received through the first input end of the first AND gate D1, and an interrupt signal sent by the mainboard management chip is received through the second input end of the first AND gate D1; a first frequency reduction signal is generated according to the first control signal and the interrupt signal through the first AND gate D1, and the first frequency reduction signal is sent to the external device through the output end of the first AND gate D1, so that the external device can operate at a reduced frequency.

[0114] Alternatively, see Figure 4 , the programmable logic device includes a first AND gate D1, a second AND gate D2 and a mixer MUX, and the number of the power supply units PSU is two; The two power supply units are connected respectively through the first input terminal and the second input terminal of the second AND gate D2 to respectively receive the power supply temperature alarm original signals of the two power supply units, and generate a power supply temperature alarm summary signal to be output to the first input terminal of the mixer MUX via the output terminal of the second AND gate D2; Receive the fault alarm signal of the mainboard management chip BMC through the control end of the mixer MUX; A default high-level signal is input through the second input terminal of the mixer MUX, so that the mixer MUX generates a power supply temperature alarm signal according to the power supply temperature alarm summary signal, the fault alarm signal and the high-level signal, and outputs the power supply temperature alarm signal to the second input terminal of the first AND gate through the output terminal of the mixer; A frequency reduction signal sent by the mainboard management chip is received through a first input end of the first AND gate D1, and an interrupt signal sent by the mainboard management chip and / or a power supply temperature alarm signal sent by the mixer is received through a second input end of the first AND gate D1; A first frequency reduction signal is generated according to the frequency reduction signal, the interrupt signal and / or the power supply temperature alarm signal through the first AND gate D1, and the first frequency reduction signal is sent to the external device via the output end of the first AND gate.

[0115] Alternatively, see Figure 4 , the power supply management chip includes a processor power supply management chip, and the programmable logic device includes a buffer MUX and a third AND gate D3; monitoring the temperature of the processing unit through the processing unit power management chip, and in response to the temperature of the processing unit exceeding a second threshold, receiving an overheating signal sent by the processing unit power management chip through the buffer; Inputting the overheat signal to the first input terminal of a third AND gate through the buffer, and inputting the first frequency reduction signal to the second input terminal of the third AND gate through the first AND gate; The second frequency reduction signal is generated according to the overheat signal and the first frequency reduction signal through the third AND gate, and the second frequency reduction signal is sent to the corresponding processing unit.

[0116] Alternatively, see Figure 5 , the programmable logic device includes a fourth AND gate D4 and a trigger Trigger; the trigger Trigger is connected to the mainboard management chip BMC, and the fourth AND gate D4 is connected to the trigger Trigger and the processor CPU respectively; When the temperature of the processor exceeds a third threshold, the processor generates an over-temperature signal; The over-temperature signal is received through the first input terminal of the fourth AND gate D4, the power-on signal sent by the processor is received through the second input terminal of the fourth AND gate, and a trigger control signal is generated according to the over-temperature signal and the power-on signal through the fourth AND gate, and sent to the trigger Trigger; A trigger generates a corresponding trigger signal according to the trigger control signal, and sends the trigger signal to the power supply management chip.

[0117] Further, see Figure 6 , the power supply management chip includes a memory power supply management chip, the programmable logic device CPLD includes a buffer Register, and the memory power supply management chip is connected to the buffer Register; The temperature of the storage unit is monitored by the storage unit power supply management chip. In response to the temperature of the storage unit exceeding a fourth threshold, the power supply management chip generates a temperature warning signal; the temperature warning signal is received by the buffer Register and sent to the processing unit, so that the processing unit generates a temperature reduction signal to the storage unit power supply management chip.

[0118] Furthermore, the memory power management chip is also used for overcurrent protection of the memory. When the current of the memory monitored by the memory power management chip is greater than the second current threshold, the memory power management chip directly sends a second overcurrent protection signal to the processor, so that the processor disconnects the power supply of the memory.

[0119] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A computer monitoring management method, It is characterized in that The computer includes a processing unit and a storage unit, and the computer monitoring and management system includes a programmable controller, which is connected to the processing unit and the storage unit respectively; The method is used for a programmable controller and comprises: When the temperature of the processing unit exceeds a first threshold, generating a first control signal and sending the first control signal to an external device so that the external device can operate at a lower frequency; When the temperature of the processing unit exceeds a second threshold, generating a second frequency reduction signal and sending it to the processing unit, so that the processing unit actively performs a frequency reduction operation; When the temperature of the processing unit exceeds a third threshold, receiving an over-temperature signal sent by the processing unit, and shutting down the power supply of the processing unit; wherein the first threshold, the second threshold, and the third threshold increase in sequence; When the temperature of the storage unit is monitored to exceed a fourth threshold, a temperature warning signal is sent to the processing unit, and a temperature reduction signal sent by the processing unit is received to reduce power supply to the storage unit.

2. According to the computer monitoring management method of claim 1, the programmable controller comprises a management chip, a power supply management chip and a mainboard management chip; the processing unit, the power supply management chip and the mainboard management chip are respectively connected to the management chip; the storage unit is respectively connected to the power supply management chip and the processing unit; The method is used for the management chip, and when the temperature of the monitoring processing unit exceeds a first threshold, generates a first control signal and sends it to an external device, so that the external device reduces the frequency of operation, specifically include: When the temperature of the processing unit monitored by the motherboard management chip exceeds a first threshold, receiving a first control signal sent by the motherboard management chip, generating a first frequency reduction signal and sending it to an external device, so that the external device can operate at a reduced frequency; When the temperature of the processing unit exceeds a second threshold, generating a second frequency reduction signal and sending it to the processing unit so that the processing unit actively performs a frequency reduction operation, specifically comprising: When the temperature of the processing unit monitored by the power management chip exceeds a second threshold, an overheating signal sent by the power management chip is received, and a second frequency reduction signal is generated and sent to the processing unit, so that the processing unit actively performs a frequency reduction operation; When the temperature of the processing unit exceeds a third threshold, receiving an over-temperature signal sent by the processing unit and shutting down the power supply of the processing unit specifically includes: When the temperature of the processing unit exceeds a third threshold, an over-temperature signal sent by the processing unit is received, and a second control signal is generated and sent to the power supply management chip, so that the power supply management chip generates an enable signal to shut down the power supply of the processing unit; When the temperature of the storage unit exceeds a fourth threshold, a temperature warning signal is sent to the processing unit, and a temperature reduction signal sent by the processing unit is received to reduce the power supply of the storage unit, specifically including: When the power management chip monitors that the temperature of the storage unit exceeds a fourth threshold, a temperature warning signal issued by the power management chip is received and sent to the processing unit, so that the processing unit generates a temperature reduction signal to the power management chip to reduce the power supply of the storage unit.

3. The computer monitoring management method according to claim 2, It is characterized in that The management chip is further connected to a power supply unit; Before generating the first frequency reduction signal and sending it to the processing unit, the method further includes: When the temperature of the power supply unit exceeds a fifth threshold, receiving a power supply temperature alarm signal sent by the power supply unit; Generating a first frequency reduction signal and sending it to an external device specifically includes: The first frequency reduction signal is generated according to the power supply temperature alarm signal and / or the overheating signal and sent to the external device.

4. The computer monitoring management method according to claim 2, It is characterized in that The management chip includes a first AND gate; Receiving a first control signal sent by the mainboard management chip, generating a first frequency reduction signal and sending it to an external device, specifically includes: Receiving a first control signal sent by the mainboard management chip through a first input end of the first AND gate; Receiving an interrupt signal sent by the mainboard management chip through the second input end of the first AND gate; A first frequency reduction signal is generated according to the first control signal and the interrupt signal through the first AND gate, and the first frequency reduction signal is sent to the external device via the output end of the first AND gate.

5. The computer monitoring management method according to claim 3, It is characterized in that The management chip includes a first AND gate, a second AND gate and a mixer, and there are two power supply units; Receiving a power supply temperature alarm signal sent by the power supply unit specifically includes: The two power supply units are connected respectively through the first input terminal and the second input terminal of the second AND gate to respectively receive the power supply temperature alarm original signals of the two power supply units, and generate a power supply temperature alarm summary signal to be output to the first input terminal of the mixer via the output terminal of the second AND gate; Generating the frequency reduction signal according to the power supply temperature alarm signal and / or the overheat signal and sending it to the external device specifically includes: Receiving a fault alarm signal of the mainboard management chip through the control end of the mixer; Input a default high-level signal through the second input terminal of the mixer, so that the mixer generates a power supply temperature alarm signal according to the power supply temperature alarm summary signal, the fault alarm signal and the high-level signal, and outputs the power supply temperature alarm signal to the second input terminal of the first AND gate through the output terminal of the mixer; Receiving a frequency reduction signal sent by the mainboard management chip BMC through the first input end of the first AND gate; receiving an interrupt signal sent by the mainboard management chip BMC and / or a power supply temperature alarm signal sent by the mixer through the second input end of the first AND gate; A first frequency reduction signal is generated by the first AND gate according to the frequency reduction signal, the interrupt signal and / or the power supply temperature alarm signal, and the first frequency reduction signal is sent to the external device via the output end of the first AND gate.

6. The computer monitoring management method according to claim 4 or 5, It is characterized in that The power supply management chip includes a processing unit power supply management chip, and the management chip includes a buffer and a third AND gate; When the temperature of the processing unit monitored by the power management chip exceeds a second threshold, receiving an overheating signal sent by the power management chip, and generating a second frequency reduction signal to send to the processing unit specifically includes: When the temperature of the processing unit monitored by the processing unit power supply management chip exceeds a second threshold, receiving an overheating signal sent by the processing unit power supply management chip through the buffer; Inputting the overheat signal to the first input terminal of a third AND gate through the buffer, and inputting the first frequency reduction signal to the second input terminal of the third AND gate through the first AND gate; The second frequency reduction signal is generated according to the overheat signal and the first frequency reduction signal through the third AND gate, and the second frequency reduction signal is sent to the corresponding processing unit.

7. The computer monitoring management method according to claim 6, It is characterized in that The processing unit power supply management chip is also used to perform overcurrent protection on the processing unit; The method further comprises: When the current of the processing unit monitored by the processing unit power supply management chip is greater than the first current threshold, the processing unit power supply management chip directly sends a first overcurrent protection signal to the processing unit to stop the processing unit and cut off the power supply.

8. The computer monitoring management method according to claim 2, It is characterized in that The management chip includes a fourth AND gate and a trigger; the trigger is connected to the mainboard management chip, and the fourth AND gate is connected to the trigger and the processing unit respectively; When the temperature of the processing unit exceeds a third threshold, receiving an over-temperature signal sent by the processing unit, and generating a second control signal to send to the power management chip, specifically includes: When the temperature of the processing unit exceeds a third threshold, receiving an over-temperature signal sent by the processing unit through the first input terminal of the fourth AND gate; receiving a power-on signal sent by the processing unit through the second input terminal of the fourth AND gate; Generate a first frequency reduction signal according to the over-temperature signal and the power-on signal through the fourth AND gate, and send the signal to the trigger; A corresponding trigger signal is generated according to the first frequency reduction signal by the trigger, and the trigger signal is sent to the power supply management chip.

9. The computer monitoring management method according to claim 8, It is characterized in that The management chip includes a buffer, and the buffer is connected to the mainboard management chip; When the temperature of the processing unit exceeds a third threshold, the method further includes: The over-temperature signal sent by the processing unit is received through the buffer, so that the mainboard management chip reads the over-temperature signal and issues a warning.

10. The computer monitoring management method according to claim 2, It is characterized in that The power supply management chip includes a storage unit power supply management chip, the management chip includes a buffer, and the storage unit power supply management chip is connected to the buffer; When the temperature of the storage unit monitored by the power management chip exceeds a fourth threshold, receiving a temperature warning signal issued by the power management chip and sending the signal to the processing unit, so that the processing unit generates a temperature reduction signal to the power management chip, specifically comprising: When the storage unit power supply management chip monitors that the temperature of the storage unit exceeds a fourth threshold, the buffer receives a temperature warning signal issued by the storage unit power supply management chip and sends it to the processing unit, so that the processing unit generates a temperature reduction signal to the storage unit power supply management chip.

11. The computer monitoring management method according to claim 10, It is characterized in that The storage unit power supply management chip is also used for overcurrent protection of the storage unit; The method also includes: when the current of the storage unit monitored by the storage unit power supply management chip is greater than a second current threshold, directly sending a second overcurrent protection signal to the processing unit through the storage unit power supply management chip to cause the processing unit to disconnect the power supply of the storage unit.

12. A computer monitoring and management system, the computer comprising a processing unit and a storage unit, It is characterized in that include: A programmable controller, wherein the programmable controller is connected to the processing unit and the storage unit respectively, and is configured as follows: In response to the temperature of the processing unit exceeding a first threshold, generating a first control signal and sending the first control signal to an external device, so that the external device operates at a reduced frequency; In response to the temperature of the processing unit exceeding a second threshold, generating a second frequency reduction signal and sending it to the processing unit, so that the processing unit actively performs a frequency reduction operation; In response to an over-temperature signal sent by the processing unit, turning off the power supply of the processing unit; wherein the over-temperature signal sent by the processing unit is generated when the temperature of the processing unit exceeds a third threshold, and the first threshold, the second threshold and the third threshold increase in sequence; In response to monitoring that the temperature of the storage unit exceeds a fourth threshold, a temperature warning signal is sent to the processing unit, and a temperature reduction signal sent by the processing unit is received to reduce the power supply of the storage unit.

13. The computer monitoring management system according to claim 12, It is characterized in that The programmable controller includes a management chip, a power supply management chip and a mainboard management chip; the processing unit, the power supply management chip and the mainboard management chip are respectively connected to the management chip; the storage unit is respectively connected to the power supply management chip and the processing unit; The temperature of the processing unit is monitored by the mainboard management chip, and in response to the temperature of the processing unit exceeding a first threshold, a first control signal sent by the mainboard management chip is received by the management chip, a first frequency reduction signal is generated and sent to an external device, so that the external device can operate at a reduced frequency; The power supply management chip monitors the temperature of the processing unit, and in response to the temperature of the processing unit exceeding a second threshold, the management chip receives an overheating signal sent by the power supply management chip, generates a second frequency reduction signal and sends it to the processing unit, so that the processing unit actively performs a frequency reduction operation; In response to receiving the over-temperature signal sent by the processing unit, generating a second control signal through the management chip and sending it to the power management chip, so that the power management chip generates an enable signal to turn off the power supply of the processing unit; The temperature of the storage unit is monitored by the power supply management chip. In response to exceeding a fourth threshold, a temperature warning signal issued by the power supply management chip is received by the management chip and sent to the processing unit, so that the processing unit generates a temperature reduction signal to the power supply management chip to reduce power supply to the storage unit.

14. The computer monitoring management system according to claim 13, It is characterized in that The management chip is further connected to a power supply unit; In response to the temperature of the power supply unit exceeding the fifth threshold, the management chip receives a power temperature alarm signal sent by the power supply unit, and generates the first frequency reduction signal based on the power temperature alarm signal and / or the overheating signal and sends it to the external device.

15. The computer monitoring management system according to claim 13, It is characterized in that The management chip includes a first AND gate; A first control signal sent by the mainboard management chip is received through the first input end of the first AND gate, and an interrupt signal sent by the mainboard management chip is received through the second input end of the first AND gate; a first frequency reduction signal is generated according to the first control signal and the interrupt signal through the first AND gate, and the first frequency reduction signal is sent to the external device via the output end of the first AND gate, so that the external device can operate at a reduced frequency.

16. The computer monitoring management system according to claim 14, It is characterized in that The management chip includes a first AND gate, a second AND gate and a mixer, and there are two power supply units; The two power supply units are connected respectively through the first input terminal and the second input terminal of the second AND gate to respectively receive the power supply temperature alarm original signals of the two power supply units, and generate a power supply temperature alarm summary signal to be output to the first input terminal of the mixer via the output terminal of the second AND gate; Receiving a fault alarm signal of the mainboard management chip through the control end of the mixer; Input a default high-level signal through the second input terminal of the mixer, so that the mixer generates a power supply temperature alarm signal according to the power supply temperature alarm summary signal, the fault alarm signal and the high-level signal, and outputs the power supply temperature alarm signal to the second input terminal of the first AND gate through the output terminal of the mixer; Receiving a frequency reduction signal sent by the mainboard management chip through a first input end of the first AND gate, and receiving an interrupt signal sent by the mainboard management chip and / or a power supply temperature alarm signal sent by the mixer through a second input end of the first AND gate; A first frequency reduction signal is generated by the first AND gate according to the frequency reduction signal, the interrupt signal and / or the power supply temperature alarm signal, and the first frequency reduction signal is sent to the external device via the output end of the first AND gate.

17. The computer monitoring management system according to claim 15 or 16, It is characterized in that The power supply management chip includes a processing unit power supply management chip, and the management chip includes a buffer and a third AND gate; monitoring the temperature of the processing unit through the processing unit power management chip, and in response to the temperature of the processing unit exceeding a second threshold, receiving an overheating signal sent by the processing unit power management chip through the buffer; Inputting the overheat signal to the first input terminal of a third AND gate through the buffer, and inputting the first frequency reduction signal to the second input terminal of the third AND gate through the first AND gate; The second frequency reduction signal is generated according to the overheat signal and the first frequency reduction signal through the third AND gate, and the second frequency reduction signal is sent to the corresponding processing unit.

18. The computer monitoring management system according to claim 13, It is characterized in that The management chip includes a fourth AND gate and a trigger; the trigger is connected to the mainboard management chip, and the fourth AND gate is connected to the trigger and the processing unit respectively; Receiving the over-temperature signal through the first input terminal of the fourth AND gate, receiving the power-on signal sent by the processing unit through the second input terminal of the fourth AND gate, generating a first frequency reduction signal according to the over-temperature signal and the power-on signal through the fourth AND gate, and sending the first frequency reduction signal to the trigger; A corresponding trigger signal is generated according to the first frequency reduction signal by the trigger, and the trigger signal is sent to the power supply management chip.

19. The computer monitoring management system according to claim 13, It is characterized in that The power supply management chip includes a storage unit power supply management chip, the management chip includes a buffer, and the storage unit power supply management chip is connected to the buffer; The power supply management chip of the storage unit monitors the temperature of the storage unit, and in response to the temperature of the storage unit exceeding a fourth threshold, the power supply management chip generates a temperature warning signal; The temperature warning signal is received by the buffer and sent to the processing unit, so that the processing unit generates a temperature reduction signal to the storage unit power supply management chip.

Citation Information

Patent Citations

  • Performance and power consumption management system and method for multi-core heterogeneous chip

    CN114637387A

  • Power supply control system and method for preventing server from being powered off and shut down

    CN115904050A

  • Temperature management system and temperature management method based on temperature management system

    CN116107406A

  • Mainboard protection system and method

    WO2024103745A1