WO2020258569A1 - 一种整机箱功耗的分配方法、系统、装置及可读存储介质 - Google Patents

一种整机箱功耗的分配方法、系统、装置及可读存储介质 Download PDF

Info

Publication number
WO2020258569A1
WO2020258569A1 PCT/CN2019/108980 CN2019108980W WO2020258569A1 WO 2020258569 A1 WO2020258569 A1 WO 2020258569A1 CN 2019108980 W CN2019108980 W CN 2019108980W WO 2020258569 A1 WO2020258569 A1 WO 2020258569A1
Authority
WO
WIPO (PCT)
Prior art keywords
power consumption
node
preset
total amount
standby
Prior art date
Application number
PCT/CN2019/108980
Other languages
English (en)
French (fr)
Inventor
韩红瑞
Original Assignee
苏州浪潮智能科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 苏州浪潮智能科技有限公司 filed Critical 苏州浪潮智能科技有限公司
Priority to US17/281,590 priority Critical patent/US11586264B2/en
Publication of WO2020258569A1 publication Critical patent/WO2020258569A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
    • G06F9/5022—Mechanisms to release resources
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
    • G06F1/26—Power supply means, e.g. regulation thereof
    • G06F1/28—Supervision thereof, e.g. detecting power-supply failure by out of limits supervision
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
    • G06F1/26—Power supply means, e.g. regulation thereof
    • G06F1/32—Means for saving power
    • G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
    • G06F1/3206—Monitoring of events, devices or parameters that trigger a change in power modality
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5094—Allocation of resources, e.g. of the central processing unit [CPU] where the allocation takes into account power or heat criteria
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Definitions

  • the present invention relates to the field of energy consumption in a computer room, in particular to a method, system, device and readable storage medium for allocating power consumption of a whole chassis.
  • the usual restriction methods include manually determining the power consumption limit of each node.
  • the power consumption of each node is independent of each other, and the total power consumption does not exceed the limit; or through the CMC (Chassis Management Controller) management on the entire chassis
  • the module sets the power consumption of the entire chassis, and then the CMC evenly distributes the total power consumption to each node.
  • the purpose of the present invention is to provide a method, system, device and readable storage medium for allocating power consumption of the entire chassis.
  • the specific plan is as follows:
  • a method for allocating power consumption of the entire chassis, applied to CMC includes:
  • the total amount of standby power consumption and the preset power consumption of the node Power consumption is redistributed.
  • the preset power consumption corresponding to each node is specifically the preset power consumption determined according to the node level of the node.
  • the process of redistributing the preset power consumption includes:
  • the first rate is determined according to the increase rate. 1. Allocating power consumption, allocating the first allocated power consumption from the total amount of standby power consumption to the preset power consumption of the node;
  • the second allocated power consumption is to allocate the second allocated power consumption from the preset power consumption of the node to the total amount of standby power consumption.
  • the first utilization rate is not less than the second utilization rate.
  • the rate of change corresponding to the actual power consumption of the node is specifically the rate of change of power consumption utilization.
  • the allocation method further includes:
  • the power consumption recovery condition is specifically that the power consumption utilization rate of the node is lower than the third utilization rate
  • the process of reallocating the preset power consumption and the total amount of standby power consumption of any one of the power consumption to-be-recovered nodes specifically includes:
  • the allocation method further includes:
  • the absolute value ratio of the total amount of standby power consumption to the total amount of remaining power consumption is used to calculate the The remaining power consumption is allocated to the total amount of standby power consumption;
  • the rated power consumption of the entire chassis is allocated to each node according to the ratio of the actual power consumption of all the nodes The preset power consumption.
  • the present invention also discloses a distribution system for power consumption of the entire chassis, which is applied to CMC, and includes:
  • the standby determining module is used to determine the total standby power consumption according to the rated power consumption of the entire chassis and the preset power consumption of each node;
  • the first allocation module is configured to allocate corresponding preset power consumption to each of the nodes
  • the power consumption detection module is used to detect the actual power consumption of each node according to the detection frequency
  • the second allocation module is used to calculate the total standby power consumption according to the change speed of the actual power consumption of each node and/or the power utilization rate of the actual power consumption of the node in the preset power consumption of the node. The amount is redistributed with the preset power consumption of the node.
  • the present invention discloses a device for distributing power consumption of the entire chassis, which is applied to CMC, and includes:
  • Memory used to store computer programs
  • the processor is used to implement the steps of the method for allocating power consumption of the entire chassis as described above when executing the computer program.
  • the present invention discloses a readable storage medium, which is applied to CMC, and a computer program is stored on the readable storage medium.
  • the computer program is executed by a processor, the power consumption of the entire chassis is realized as described above. The steps of the allocation method.
  • the present invention is applied to the case that the rated power consumption of the entire chassis is limited, the preset power consumption of each node is automatically adjusted by the CMC, and the actual power consumption change and the power consumption utilization rate of each node detected are readjusted.
  • the preset power consumption of the node so as to ensure that the node can have a larger power consumption support when the business is busy, process the business with better performance, and release the idle power consumption to the total standby power consumption when the business is small, avoiding excessive Occupation of power consumption causes the processing efficiency of other nodes to be restricted by power consumption, which maximizes the effective utilization of the entire chassis, reduces power consumption and vacancy, and reduces operating costs.
  • Figure 1 is a flow chart of the steps of a method for allocating power consumption of the entire chassis in an embodiment of the present invention
  • Figure 2 is a structural topology diagram of the internal equipment of the entire chassis in an embodiment of the present invention.
  • Figure 3 is a structural distribution diagram of a distribution system for power consumption of the entire chassis in an embodiment of the present invention
  • Fig. 4 is a structural distribution diagram of a device for distributing power consumption of the entire chassis in an embodiment of the present invention.
  • the business workload of its nodes is not consistent.
  • power consumption limits the processing speed and efficiency of nodes, causing the business processing volume to continue to rise, while nodes with idle business have functions.
  • the situation of wasteful power consumption results in idle power consumption and inability to make full use of the total power consumption of the chassis, resulting in a large amount of waste of power resources.
  • the CMC automatically adjusts the preset power consumption of each node, and readjusts the preset power consumption of the node according to the detected actual power consumption change and power consumption utilization rate of each node to avoid excessive power consumption.
  • the processing efficiency of other nodes is limited by power consumption, which maximizes the effective utilization of the entire chassis, reduces power consumption and vacancy, and reduces operating costs.
  • the embodiment of the present invention discloses a method for allocating power consumption of the entire chassis, which is applied to CMC, as shown in Fig. 1, including:
  • S1 Determine the total amount of standby power consumption according to the rated power consumption of the entire chassis and the preset power consumption of each node;
  • the preset power consumption of all nodes is subtracted from the rated power consumption of the entire chassis to obtain the total standby power consumption.
  • the preset power consumption corresponding to each node is generally the preset power consumption determined according to the node level of the node. The higher the node level, the greater the corresponding preset power consumption.
  • the CMC in the entire chassis allocates the preset power consumption to each node, which actually sends the allocation to the BMC (Baseboard Management Controller) in each node.
  • the configuration information of the preset power consumption, the SYSTEM of each node determines the system operation strategy according to the configuration information of the preset power consumption in the BMC.
  • the SYSTEM of each node is powered by the PSU (Power Supply Unit) of the node where it is located. Therefore, the detection object in step S3 is actually the total power consumption of all PSUs in each node.
  • the middle BMC accumulates the power consumption of all corresponding PSUs and provides it for CMC to query.
  • the CMC extracts real-time power consumption data in each BMC according to the detection frequency, which can be set by the user in the CMC.
  • the process of reallocating preset power consumption includes:
  • the first rate is determined according to the increase rate. 1. Allocating power consumption, allocating the first allocated power consumption from the total amount of standby power consumption to the preset power consumption of the node;
  • the second allocated power consumption is to allocate the second allocated power consumption from the preset power consumption of the node to the total amount of standby power consumption.
  • the change speed of the actual power consumption of the node may be the change amount of the actual power consumption of the node obtained by two detection actions, or the change speed of the power consumption utilization rate.
  • the second utilization rate can be set to 50%, and the second distribution power consumption is determined according to the decreasing speed, and the second distribution power consumption is allocated from the preset power consumption of the node After the process of reaching the total amount of standby power consumption, it can achieve the purpose of a higher power consumption utilization rate of its actual power consumption accounting for the preset power consumption.
  • the power consumption utilization rate needs to reach more than 80%.
  • first speed, second speed, first utilization rate, second utilization rate, first distribution power consumption, and second distribution power consumption mentioned above are all positive values.
  • the specific characteristic values of the power consumption utilization rate, such as the first utilization rate, the second utilization rate, etc. only need to meet the corresponding service requirements, and there is no specific numerical limit.
  • the first utilization rate is usually not less than the second utilization rate.
  • the invention discloses a method for allocating the power consumption of the entire chassis, which is applied to CMC, including: determining the total amount of standby power consumption according to the rated power consumption of the entire chassis and the preset power consumption of each node; Allocate the corresponding preset power consumption; detect the actual power consumption of each node according to the detection frequency; according to the change speed of the actual power consumption of each node and/or the actual power consumption of the node accounts for the preset of the node
  • the power consumption utilization rate of power consumption redistributes the total amount of standby power consumption and the preset power consumption of the node.
  • the present invention is applied to the case that the rated power consumption of the entire chassis is limited, the preset power consumption of each node is automatically adjusted by the CMC, and the actual power consumption change and the power consumption utilization rate of each node detected are readjusted.
  • the preset power consumption of the node so as to ensure that the node can have a larger power consumption support when the business is busy, process the business with better performance, and release the idle power consumption to the total standby power consumption when the business is small, avoiding excessive Occupation of power consumption causes the processing efficiency of other nodes to be restricted by power consumption, which maximizes the effective utilization of the entire chassis, reduces power consumption and vacancy, and reduces operating costs.
  • the embodiment of the present invention discloses a specific method for allocating the power consumption of the entire chassis. Compared with the previous embodiment, this embodiment further illustrates and optimizes the technical solution. Specifically, the allocation method further includes:
  • the power consumption recovery condition is specifically that the power consumption utilization rate of the node is lower than the third utilization rate
  • the process of reallocating the preset power consumption and the total amount of standby power consumption of any one of the power consumption to-be-recovered nodes specifically includes:
  • the third utilization rate and the fourth utilization rate are 80% and 90%, respectively, and determine the node whose power consumption utilization rate is less than 80% as the node whose power consumption is to be recovered, and collect it from the preset power consumption of the node
  • the third allocation of power consumption is allocated to the total standby power consumption.
  • the power consumption utilization rate of a node is 75%, and a part of the power consumption value is recovered in the total standby power consumption so that the power consumption utilization rate of the node exceeds 90%.
  • the node level can also be considered in the process of power recovery.
  • the lower the node level the higher the third utilization rate and the easier it is to be determined as the power recovery node. It is assumed that the faster the power consumption is recovered by the CMC to the total standby power consumption.
  • the rated power consumption of the entire chassis in the CMC in this embodiment may also change.
  • the preset power consumption and standby power consumption of the nodes inside the entire chassis need to be adjusted accordingly. Total. If the total standby power consumption at this time is not less than the change in the rated power consumption of the entire chassis, the relative power consumption in the total standby power consumption can be directly cut off. If the total standby power consumption is less than the rated power consumption of the entire chassis The amount of change in the standby power consumption is negative after subtracting the amount of change.
  • the allocation method further includes:
  • the absolute value ratio of the total amount of standby power consumption to the total amount of remaining power consumption is used to calculate the The remaining power consumption is allocated to the total amount of standby power consumption;
  • the rated power consumption of the entire chassis is allocated to each node according to the ratio of the actual power consumption of all the nodes The preset power consumption.
  • the original whole chassis runs at a rated power consumption of 2000W, and then it is reduced to 1500W with a change of 500W.
  • the preset power consumption and actual power consumption of all nodes in the chassis are calculated separately to obtain the remaining power consumption, and all remaining power consumption is summed to obtain the total remaining power consumption;
  • the total remaining power consumption is 400, that is to say, the total remaining power consumption and the total standby power consumption are positive, then according to the absolute value ratio of the total standby power consumption to the total remaining power consumption 3:4, the 3/4 of the remaining power in each node is allocated to the total standby power consumption to fill the gap of the total standby power consumption; assuming that the total remaining power consumption is 200, that is, the total remaining power consumption is equal to the total standby power consumption If the value is negative, the current rated power consumption of 1500W is allocated to the preset power consumption of each node directly according to the ratio of the actual power consumption at the time when each node is running. At this time, the total preset power consumption is zero.
  • the SYSTEM of the node determines the corresponding power supply operation strategy according to the corresponding preset power consumption.
  • the rated power consumption of the entire chassis in the CMC in this embodiment changes, and the user increases the rated power consumption of the original entire chassis, the increased rated power can be directly added to the total standby power consumption. .
  • the CMC can dynamically adjust the actual power consumption of each node in the chassis according to the needs of their respective business volume.
  • the node power consumption can be automatically reduced by means of frequency reduction, and the heat can be effectively reduced.
  • Useless power consumption improves node performance; when the node's business volume increases, it can add power allocation in time to improve business processing capabilities and improve node performance.
  • the embodiment of the present invention also discloses a distribution system for power consumption of the entire chassis, which is applied to CMC, as shown in FIG. 3, including:
  • the standby determination module 01 is used to determine the total standby power consumption according to the rated power consumption of the entire chassis and the preset power consumption of each node;
  • the first allocation module 02 is configured to allocate corresponding preset power consumption to each of the nodes
  • the power consumption detection module 03 is configured to detect the actual power consumption of each node according to the detection frequency
  • the second allocation module 04 is configured to calculate the standby power consumption according to the change speed of the actual power consumption of each node and/or the power consumption utilization rate of the actual power consumption of the node in the preset power consumption of the node. The total amount is redistributed with the preset power consumption of the node.
  • the embodiment of the present invention is applied to the case that the rated power consumption of the entire chassis is limited, the CMC automatically adjusts the preset power consumption of each node, and re-applies the detected actual power consumption change and power consumption utilization rate of each node. Adjust the preset power consumption of the node, so as to ensure that the node can have a larger power consumption support when the business is busy, process the business with better performance, and release the idle power consumption to the total standby power consumption when the business is less. Excessive power consumption causes the processing efficiency of other nodes to be restricted by power consumption, which maximizes the effective utilization of the entire chassis, reduces power consumption and vacancy, and reduces operating costs.
  • the preset power consumption corresponding to each node is specifically the preset power consumption determined according to the node level of the node.
  • the second allocation module 04 determines the power consumption rate of any node according to the change speed of the actual power consumption and/or the actual power consumption of the node in the preset power consumption of the node.
  • the process of reallocating the total amount of standby power consumption and the preset power consumption of the node specifically includes: when the actual power consumption of the node increases faster than the first speed and/or the actual power consumption of the node accounts for the node If the power consumption utilization rate of the preset power consumption exceeds the first utilization rate, the first allocated power consumption is determined according to the increase speed, and the first allocated power consumption is allocated to the node from the total amount of standby power consumption In the preset power consumption; when the reduction rate of the actual power consumption of the node exceeds the second speed and/or the actual power consumption of the node accounts for the preset power consumption of the node, the power consumption utilization rate is lower than the second utilization rate, Then, the second allocated power consumption is determined according to the reduction speed, and the second allocated power consumption is allocated from the preset power consumption of
  • the first utilization rate is not less than the second utilization rate.
  • the rate of change corresponding to the actual power consumption of the node is specifically the rate of change of power consumption utilization.
  • the distribution system for power consumption of the entire chassis further includes a third distribution module, configured to: when the total amount of standby power consumption is zero, confirm that the node that meets the power recovery condition is power consumption The node to be recycled redistributes the preset power consumption of the node to be recycled and the total amount of standby power consumption.
  • the power consumption recovery condition is specifically that the power consumption utilization rate of the node is lower than the third utilization rate
  • the process of the third allocation module redistributing the preset power consumption of any node to be recovered and the total amount of standby power consumption specifically includes:
  • the distribution system for the power consumption of the entire chassis further includes a fourth distribution module, which is used for when the total amount of standby power consumption is negative, the preset power consumption and the corresponding actual power consumption of all the nodes The difference of power consumption is calculated respectively to obtain the remaining power consumption, and all the remaining power consumptions are summed to obtain the total remaining power consumption; when the total amount of standby power consumption and the total remaining power consumption are positive, then According to the absolute value ratio of the total standby power consumption to the total remaining power consumption, the remaining power consumption of each node is allocated to the total standby power consumption; when the total standby power consumption The total amount of remaining power consumption is negative, and the rated power consumption of the entire chassis is allocated to the preset power consumption of each node according to the ratio of the actual power consumption of all the nodes.
  • a fourth distribution module which is used for when the total amount of standby power consumption is negative, the preset power consumption and the corresponding actual power consumption of all the nodes The difference of power consumption is calculated respectively to obtain the remaining power consumption, and all
  • the embodiment of the present application also discloses a device for distributing power consumption of the entire chassis, which is applied to the CMC. As shown in FIG. 4, it includes a processor 11 and a memory 12; wherein, the processing 11 executes the data stored in the memory 12
  • the computer program implements the following steps:
  • the total amount of standby power consumption and the preset power consumption of the node Power consumption is redistributed.
  • the embodiment of the present invention is applied to the case that the rated power consumption of the entire chassis is limited, the CMC automatically adjusts the preset power consumption of each node, and re-starts the detection according to the actual power consumption change and power consumption utilization rate of each node detected Adjust the preset power consumption of the node, so as to ensure that the node can have a larger power consumption support when the business is busy, process the business with better performance, and release the idle power consumption to the total standby power consumption when the business is less.
  • Excessive power consumption causes the processing efficiency of other nodes to be restricted by power consumption, which maximizes the effective utilization of the entire chassis, reduces power consumption and vacancy, and reduces operating costs.
  • the preset power consumption corresponding to each node is specifically the preset power consumption determined according to the node level of the node.
  • the first rate is determined according to the increase rate. 1. Allocating power consumption, allocating the first allocated power consumption from the total amount of standby power consumption to the preset power consumption of the node;
  • the second allocated power consumption is to allocate the second allocated power consumption from the preset power consumption of the node to the total amount of standby power consumption.
  • the first utilization rate is not less than the second utilization rate.
  • the rate of change corresponding to the actual power consumption of the node is specifically the rate of change of power consumption utilization.
  • the power consumption recovery condition is specifically that the power consumption utilization rate of the node is lower than the third utilization rate
  • the process of reallocating the preset power consumption and the total amount of standby power consumption of any one of the power consumption to-be-recovered nodes specifically includes:
  • the absolute value ratio of the total amount of standby power consumption to the total amount of remaining power consumption is used to calculate the The remaining power consumption is allocated to the total amount of standby power consumption;
  • the rated power consumption of the entire chassis is allocated to the preset power consumption of each node according to the ratio of the actual power consumption of all the nodes. Set power consumption.
  • the device for distributing power consumption of the entire chassis in this embodiment may further include:
  • the input interface 13 is used to obtain computer programs imported from the outside world, and save the obtained computer programs in the memory 12, and can also be used to obtain various instructions and parameters transmitted by external terminal devices, such as preset power consumption and detection. Frequency, the actual power consumption of the BMC transmission of each node is obtained, and transmitted to the processor 11, so that the processor 11 uses the above-mentioned various instructions and parameters to carry out corresponding processing.
  • the input interface 13 may specifically include, but is not limited to, a USB interface, a serial interface, a voice input interface, a LAN interface, an I2C interface, and a hard disk reading interface.
  • the output interface 14 is used to output various data generated by the processor 11 to the terminal device connected to it, so that other terminal devices connected to the output interface 14 can obtain various data generated by the processor 11, such as redistributing
  • the preset power consumption information is sent to the corresponding nodes.
  • the output interface 14 may specifically include but is not limited to a LAN interface, an I2C interface, a USB interface, a serial interface, and the like.
  • the communication unit 15 is used to establish a remote communication connection between the distribution device and an external device such as a server or a client, so that the external device can view the internal status information of the distribution device and send parameter commands to the distribution device.
  • the communication unit 15 may specifically include, but is not limited to, a remote communication unit based on wireless communication technology or wired communication technology.
  • the keyboard 16 is used to obtain various parameter data or instructions input by the user by tapping the keycap in real time.
  • the display 17 is used for real-time display of relevant information about the power consumption distribution process, so that the user can understand the current distribution of the power consumption of the entire chassis in time.
  • the mouse 18 can be used to assist the user in entering data and simplify the user's operation.
  • the embodiment of the present application also discloses a computer-readable storage medium.
  • the computer-readable storage medium mentioned here includes random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, and Erase programmable ROM, register, hard disk, removable hard disk, CD-ROM or any other form of storage medium known in the technical field.
  • RAM random access memory
  • ROM read-only memory
  • Erase programmable ROM register, hard disk, removable hard disk, CD-ROM or any other form of storage medium known in the technical field.
  • a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the following steps are implemented:
  • the total amount of standby power consumption and the preset power consumption of the node Power consumption is redistributed.
  • the embodiment of the present invention is applied to the case that the rated power consumption of the entire chassis is limited, the CMC automatically adjusts the preset power consumption of each node, and re-applies the detected actual power consumption change and power consumption utilization rate of each node. Adjust the preset power consumption of the node, so as to ensure that the node can have a larger power consumption support when the business is busy, process the business with better performance, and release the idle power consumption to the total standby power consumption when the business is less. Excessive power consumption causes the processing efficiency of other nodes to be restricted by power consumption, which maximizes the effective utilization of the entire chassis, reduces power consumption and vacancy, and reduces operating costs.
  • the preset power consumption corresponding to each node is specifically the preset power consumption determined according to the node level of the node.
  • the first rate is determined according to the increase rate. 1. Allocating power consumption, allocating the first allocated power consumption from the total amount of standby power consumption to the preset power consumption of the node;
  • the second allocated power consumption is to allocate the second allocated power consumption from the preset power consumption of the node to the total amount of standby power consumption.
  • the first utilization rate is not less than the second utilization rate.
  • the rate of change corresponding to the actual power consumption of the node is specifically the rate of change of power consumption utilization.
  • the power consumption recovery condition is specifically that the power consumption utilization rate of the node is lower than the third utilization rate
  • the process of reallocating the preset power consumption and the total amount of standby power consumption of any one of the power consumption to-be-recovered nodes specifically includes:
  • the absolute value ratio of the total amount of standby power consumption to the total amount of remaining power consumption is used to calculate the The remaining power consumption is allocated to the total amount of standby power consumption;
  • the rated power consumption of the entire chassis is allocated to the preset power consumption of each node according to the ratio of the actual power consumption of all the nodes. Set power consumption.
  • angles ⁇ , ⁇ , ⁇ , ⁇ , ⁇ , ⁇ , ⁇ , ⁇ , ⁇ in the present invention are also distinctions made for convenience of explanation, and the symbols themselves do not have any meaning.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Supply And Distribution Of Alternating Current (AREA)
  • Remote Monitoring And Control Of Power-Distribution Networks (AREA)

Abstract

一种整机箱功耗的分配方法、系统、装置及可读存储介质，应用于CMC，包括：根据整机箱额定功耗和各节点的预设功耗，确定备用功耗总量(S1)；向每个节点分配对应的预设功耗(S2)；按照检测频率检测每个节点的实际功耗(S3)；根据每个节点的实际功耗的变化速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率，对备用功耗总量与该节点的预设功耗进行再分配(S4)。应用于整机箱额定功耗有限制的情况下，由CMC自动调节各节点的预设功耗，从而保证节点在业务繁忙时能够有较大功耗支撑，在业务较少时释放空闲功耗到备用功耗总量中，避免过多功耗占用，导致其他节点的处理效率受到功耗限制，最大限度提高了整机箱的有效利用率，运行成本降低。

Description

一种整机箱功耗的分配方法、系统、装置及可读存储介质
本申请要求于2019年06月27日提交中国专利局、申请号为201910568193.3、发明名称为“一种整机箱功耗的分配方法、系统、装置及可读存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及机房能耗领域，具体涉及一种整机箱功耗的分配方法、系统、装置及可读存储介质。
背景技术
随着现代机房的集中化和集群化，服务器节点和集群节点的数量快速增长，不可避免地带来电力消耗的快速增长，而现有机房电力改造受制于各种外部因素，未必能及时跟进，因此一些新设备更新时，需要对设备功耗记性一定的限制。
通常的限制手段，包括手动确定每个节点的功耗限额，每个节点功耗相互独立，总功耗不超限；或通过整机箱上的CMC(Chassis Management Controller，机箱管理控制器)管理模块进行整机箱的功耗设置，然后由CMC将总功耗平均分配到各个节点。
然而，无论手动设置还是CMC平均分配功耗，其节点的业务工作量并不一致，在业务繁忙的节点，功耗限制节点的处理速度和效率，使得业务处理量持续攀升，而业务空闲的节点则存在功耗浪费的情况，造成功耗闲置，无法充分利用该机箱的总功耗，造成大量电力资源的浪费。
发明内容
有鉴于此，本发明的目的在于提供一种整机箱功耗的分配方法、系统、装置及可读存储介质。其具体方案如下：
一种整机箱功耗的分配方法，应用于CMC，包括：
根据整机箱额定功耗和各节点的预设功耗，确定备用功耗总量；
向每个所述节点分配对应的预设功耗；
按照检测频率检测每个所述节点的实际功耗；
根据每个所述节点的实际功耗的变化速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率，对所述备用功耗总量与该节点的预设功耗进行再分配。
优选的，每个所述节点对应的预设功耗，具体为根据该节点的节点等级确定的预设功耗。
优选的，根据任一所述节点的实际功耗的变化速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率，对所述备用功耗总量与该节点的预设功耗进行再分配的过程，具体包括：
当该节点的实际功耗的增加速度超出第一速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率超出第一利用率，则根据所述增加速度确定第一分配功耗，将所述第一分配功耗从所述备用功耗总量中分配至该节点的预设功耗中；
当该节点的实际功耗的降低速度超出第二速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率低于第二利用率，则根据所述降低速度确定第二分配功耗，将所述第二分配功耗从该节点的预设功耗中分配至所述备用功耗总量中。
优选的，所述第一利用率不小于所述第二利用率。
优选的，所述节点的实际功耗对应的变化速度具体为功耗利用率的变化速度。
优选的，所述分配方法还包括：
当所述备用功耗总量为零，确认符合功耗回收条件的所述节点为功耗待回收节点，对所述功耗待回收节点的预设功耗和所述备用功耗总量进行再分配。
优选的，所述功耗回收条件具体为所述节点的功耗利用率低于第三利用率；
相应的，对任一所述功耗待回收节点的预设功耗和所述备用功耗总量进行再分配的过程，具体包括：
确定第三分配功耗，将所述第三分配功耗从该功耗待回收节点的预设功耗中分配至所述备用功耗总量中，同时使该功耗待回收节点的功耗利用率达到或超过第四利用率。
优选的，所述分配方法还包括：
当所述备用功耗总量为负，对所有所述节点中预设功耗和对应的实际功耗分别求差，得到剩余功耗，对所有所述剩余功耗求和，得到剩余功耗总量；
当所述备用功耗总量与所述剩余功耗总量相加为正，则按照所述备用功耗总量与所述剩余功耗总量的绝对值比，将每个所述节点的剩余功耗分配至所述备用功耗总量中；
当所述备用功耗总量与所述剩余功耗总量相加为负，则按照所有所述节点的实际功耗的比例，将所述整机箱额定功耗分配至每个所述节点的预设功耗。
相应的，本发明还公开了一种整机箱功耗的分配系统，应用于CMC，包括：
备用确定模块，用于根据整机箱额定功耗和各节点的预设功耗，确定备用功耗总量；
第一分配模块，用于向每个所述节点分配对应的预设功耗；
功耗检测模块，用于按照检测频率检测每个所述节点的实际功耗；
第二分配模块，用于根据每个所述节点的实际功耗的变化速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率，对所述备用功耗总量与该节点的预设功耗进行再分配。
相应的，本发明公开了一种整机箱功耗的分配装置，应用于CMC，包括：
存储器，用于存储计算机程序；
处理器，用于执行所述计算机程序时实现如上文所述整机箱功耗的分配方法的步骤。
相应的，本发明公开了一种可读存储介质，应用于CMC，所述可读存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现如上文 所述整机箱功耗的分配方法的步骤。
本发明应用于整机箱额定功耗有限制的情况下，由CMC自动调节各节点的预设功耗，根据检测到的每个节点的实际功耗的变化和功耗利用率，重新调整该节点的预设功耗，从而保证节点在业务繁忙时能够有较大功耗支撑，以较优的性能处理业务，在业务较少时释放空闲功耗到备用功耗总量中，避免过多功耗占用，导致其他节点的处理效率受到功耗限制，最大限度提高了整机箱的有效利用率，降低了功耗空置，运行成本降低。
附图说明
图1为本发明实施例中一种整机箱功耗的分配方法的步骤流程图；
图2为本发明实施例中一种整机箱内部设备的结构拓扑图；
图3为本发明实施例中一种整机箱功耗的分配系统的结构分布图；
图4为本发明实施例中一种整机箱功耗的分配装置的结构分布图。
具体实施方式
下面将结合本发明中的说明书附图，对发明中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
无论手动设置还是CMC平均分配功耗，其节点的业务工作量并不一致，在业务繁忙的节点，功耗限制节点的处理速度和效率，使得业务处理量持续攀升，而业务空闲的节点则存在功耗浪费的情况，造成功耗闲置，无法充分利用该机箱的总功耗，造成大量电力资源的浪费。本发明由CMC自动调节各节点的预设功耗，根据检测到的每个节点的实际功耗的变化和功耗利用率，重新调整该节点的预设功耗，避免过多功耗占用，导致其他节点的处理效率受到功耗限制，最大限度提高了整机箱的有效利用率，降低了功耗空置，运行成本降低。
本发明实施例公开了一种整机箱功耗的分配方法，应用于CMC，参见图1所示，包括：
S1：根据整机箱额定功耗和各节点的预设功耗，确定备用功耗总量；
其中，从整机箱额定功耗中减去所有节点的预设功耗，得到备用功耗总量。
每个所述节点对应的预设功耗，一般为根据该节点的节点等级确定的预设功耗，节点等级越高的节点，对应的预设功耗越大。
S2：向每个所述节点分配对应的预设功耗；
可以理解的是，参见图2的结构拓扑图，整机箱中的CMC向每个节点分配预设功耗，实际上是对每个节点中BMC(Baseboard Management Controller基板管理控制器)发送了分配预设功耗的配置信息，每个节点的SYSTEM根据BMC中的预设功耗的配置信息确定系统运行策略。
S3：按照检测频率检测每个所述节点的实际功耗；
可以理解的是，每个节点的SYSTEM由其所在节点的PSU(Power Supply Unit，供电设备)进行供电，因此步骤S3中的检测对象实际为每个节点中所有PSU的功耗总和，每个节点中BMC将对应的所有PSU的功耗累积后供CMC查询，CMC按照检测频率提取每个BMC中实时功耗的数据，该检测频率可以由用户在CMC中设定。
S4：根据每个所述节点的实际功耗的变化速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率，对所述备用功耗总量与该节点的预设功耗进行再分配。
其中，根据任一所述节点的实际功耗的变化速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率，对所述备用功耗总量与该节点的预设功耗进行再分配的过程，具体包括：
当该节点的实际功耗的增加速度超出第一速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率超出第一利用率，则根据所述增加速度确定第一分配功耗，将所述第一分配功耗从所述备用功耗总量中分配至该节点的预设功耗中；
当该节点的实际功耗的降低速度超出第二速度和/或该节点的实际功 耗占该节点的预设功耗的功耗利用率低于第二利用率，则根据所述降低速度确定第二分配功耗，将所述第二分配功耗从该节点的预设功耗中分配至所述备用功耗总量中。
可以理解的是，所述节点的实际功耗的变化速度既可以为两次检测动作得到的该节点的实际功耗的变化量，也可以是功耗利用率的变化速度。
以任一节点的实际功耗增加为例，假设其第一利用率为90％，当该节点的实际功耗占该节点的预设功耗的预设功耗率超出90％，根据历史检测数据分析当前实际功耗的增加速度是否超出第一速度，如果是，按照增加速度预测第一分配功耗，当备用功耗总量不小于第一分配功耗，将第一分配功耗从备用功耗总量中分配至该节点的预设功耗中。
任一节点的实际功耗降低与上文类似，一般可以设置第二利用率为50％，根据下降速度确定第二分配功耗，将第二分配功耗从该节点的预设功耗中分配至备用功耗总量的过程后，能够达到其实际功耗占预设功耗的功耗利用率较高的目的，一般功耗利用率需要达到80％以上。
当然，可以理解的是，上文中所述第一速度、第二速度、第一利用率、第二利用率、第一分配功耗和第二分配功耗均为正值，本实施例中有关功耗利用率的具体特征值如第一利用率、第二利用率等，满足相应的业务要求即可，没有特定的数值限定，一般来说，第一利用率通常不小于第二利用率。
本发明公开了一种整机箱功耗的分配方法，应用于CMC，包括：根据整机箱额定功耗和各节点的预设功耗，确定备用功耗总量；向每个所述节点分配对应的预设功耗；按照检测频率检测每个所述节点的实际功耗；根据每个所述节点的实际功耗的变化速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率，对所述备用功耗总量与该节点的预设功耗进行再分配。本发明应用于整机箱额定功耗有限制的情况下，由CMC自动调节各节点的预设功耗，根据检测到的每个节点的实际功耗的变化和功耗利用率，重新调整该节点的预设功耗，从而保证节点在业务繁忙时能够有较大功耗支撑，以较优的性能处理业务，在业务较少时释放空闲功耗到备用功耗总量中，避免过多功耗占用，导致其他节点的处理效率受到功耗限制， 最大限度提高了整机箱的有效利用率，降低了功耗空置，运行成本降低。
本发明实施例公开了一种具体的整机箱功耗的分配方法，相对于上一实施例，本实施例对技术方案作了进一步的说明和优化。具体的，所述分配方法还包括：
当所述备用功耗总量为零，确认符合功耗回收条件的所述节点为功耗待回收节点，对所述功耗待回收节点的预设功耗和所述备用功耗总量进行再分配。
具体的，所述功耗回收条件具体为所述节点的功耗利用率低于第三利用率；
相应的，对任一所述功耗待回收节点的预设功耗和所述备用功耗总量进行再分配的过程，具体包括：
确定第三分配功耗，将所述第三分配功耗从该功耗待回收节点的预设功耗中分配至所述备用功耗总量中，同时使该功耗待回收节点的功耗利用率达到或超过第四利用率。
例如，确定第三利用率、第四利用率分别为80％和90％，将功耗利用率低于80％的节点确定为功耗待回收节点，并从该节点的预设功耗中收取第三分配功耗分配至备用功耗总量中，如一节点的功耗利用率为75％，回收一部分功耗值备用功耗总量，以使该节点的功耗利用率超过90％。
可以理解的是，功耗回收的过程中还可以考虑节点等级，例如节点等级越低的节点，其对应的第三利用率越高，越容易被确定为功耗待回收节点，其闲置的预设功耗越快被CMC回收至备用功耗总量中。
如果备用功耗总量为零，而所有节点的功耗利用率均在第四利用率以上，则不再进行功耗再分配，由每个节点的SYSTEM自行根据预设功耗调节用电策略。
进一步的，本实施例中CMC中整机箱额定功耗也可能发生变化，当用户降低原有整机箱额定功耗，则需要相应调整整机箱内部节点的预设功耗和备用功耗总量。如果此时备用功耗总量不小于整机箱额定功耗的变化量，则直接砍去备用功耗总量中的相对功耗即可，如果备用功耗总量小于整机 箱额定功耗的变化量，则备用功耗总量减去其变化量后为负。
由此，所述分配方法还包括：
当所述备用功耗总量为负，对所有所述节点中预设功耗和对应的实际功耗分别求差，得到剩余功耗，对所有所述剩余功耗求和，得到剩余功耗总量；
当所述备用功耗总量与所述剩余功耗总量相加为正，则按照所述备用功耗总量与所述剩余功耗总量的绝对值比，将每个所述节点的剩余功耗分配至所述备用功耗总量中；
当所述备用功耗总量与所述剩余功耗总量相加为负，则按照所有所述节点的实际功耗的比例，将所述整机箱额定功耗分配至每个所述节点的预设功耗。
例如，原整机箱以额定功耗2000W运行，后被调低至1500W，变化量为500W。首先判断当前备用功耗总量是否不小于500W，如果是，直接从备用功耗总量中减去500W即可；如果否，假设减去500W后得到备用功耗总量为-300W，对整机箱内所有节点的预设功耗和实际功耗分别求差，得到剩余功耗，对所有剩余功耗求和，得到剩余功耗总量；
假设剩余功耗总量为400，也就是说剩余功耗总量与备用功耗总量相加为正，则按照备用功耗总量与剩余功耗总量的绝对值比3:4，将每个节点中3/4的剩余功率分配到备用功耗总量中，填补备用功耗总量缺损；假设剩余功耗总量为200，也即剩余功耗总量与备用功耗总量相加为负，则直接按照每个节点运行当时的实际功耗的比例，将目前1500W的额定功耗分配到每个节点的预设功耗，此时预设功耗总量为零，每个节点的SYSTEM根据对应的预设功耗确定相应的供电运行策略即可。
当然，除此之外，还可以区分不同节点的节点等级，确定不同节点的分配功耗的优先性，使节点等级高的节点更容易得到预设功耗，同时预设功耗被回收到备用功耗总量的可能性更低。
可以理解的是，如果本实施例中CMC中整机箱额定功耗发生变化，用户升高了原有整机箱额定功耗，则直接将增加的额定功率部分加入备用功耗总量即可。
通过本实施例，CMC可以时间整机箱内各节点的实际功耗根据各自业务量的需要动态调整，当节点业务量较低时，通过降频等手段自动降低节点功耗，有效降低热量等无用功耗，提高节点效能；当节点业务量上升时，又能够及时追加功率分配，提高业务处理能力，提高节点性能。
相应的，本发明实施例还公开了一种整机箱功耗的分配系统，应用于CMC，参见图3所示，包括：
备用确定模块01，用于根据整机箱额定功耗和各节点的预设功耗，确定备用功耗总量；
第一分配模块02，用于向每个所述节点分配对应的预设功耗；
功耗检测模块03，用于按照检测频率检测每个所述节点的实际功耗；
第二分配模块04，用于根据每个所述节点的实际功耗的变化速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率，对所述备用功耗总量与该节点的预设功耗进行再分配。
本发明实施例应用于整机箱额定功耗有限制的情况下，由CMC自动调节各节点的预设功耗，根据检测到的每个节点的实际功耗的变化和功耗利用率，重新调整该节点的预设功耗，从而保证节点在业务繁忙时能够有较大功耗支撑，以较优的性能处理业务，在业务较少时释放空闲功耗到备用功耗总量中，避免过多功耗占用，导致其他节点的处理效率受到功耗限制，最大限度提高了整机箱的有效利用率，降低了功耗空置，运行成本降低。
在一些具体的实施例中，每个所述节点对应的预设功耗，具体为根据该节点的节点等级确定的预设功耗。
在一些具体的实施例中，第二分配模块04根据任一所述节点的实际功耗的变化速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率，对所述备用功耗总量与该节点的预设功耗进行再分配的过程，具体包括：当该节点的实际功耗的增加速度超出第一速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率超出第一利用率，则根据所述增加速度确定第一分配功耗，将所述第一分配功耗从所述备用功耗总量中分配至该节点的预设功耗中；当该节点的实际功耗的降低速度超出第二速度和/或该节 点的实际功耗占该节点的预设功耗的功耗利用率低于第二利用率，则根据所述降低速度确定第二分配功耗，将所述第二分配功耗从该节点的预设功耗中分配至所述备用功耗总量中。
在一些具体的实施例中，所述第一利用率不小于所述第二利用率。
在一些具体的实施例中，所述节点的实际功耗对应的变化速度具体为功耗利用率的变化速度。
在一些具体的实施例中，整机箱功耗的分配系统还包括第三分配模块，用于：当所述备用功耗总量为零，确认符合功耗回收条件的所述节点为功耗待回收节点，对所述功耗待回收节点的预设功耗和所述备用功耗总量进行再分配。
在一些具体的实施例中，所述功耗回收条件具体为所述节点的功耗利用率低于第三利用率；
相应的，第三分配模块对任一所述功耗待回收节点的预设功耗和所述备用功耗总量进行再分配的过程，具体包括：
确定第三分配功耗，将所述第三分配功耗从该功耗待回收节点的预设功耗中分配至所述备用功耗总量中，同时使该功耗待回收节点的功耗利用率达到或超过第四利用率。
在一些具体的实施例中，整机箱功耗的分配系统还包括第四分配模块，用于当所述备用功耗总量为负，对所有所述节点中预设功耗和对应的实际功耗分别求差，得到剩余功耗，对所有所述剩余功耗求和，得到剩余功耗总量；当所述备用功耗总量与所述剩余功耗总量相加为正，则按照所述备用功耗总量与所述剩余功耗总量的绝对值比，将每个所述节点的剩余功耗分配至所述备用功耗总量中；当所述备用功耗总量与所述剩余功耗总量相加为负，则按照所有所述节点的实际功耗的比例，将所述整机箱额定功耗分配至每个节点的预设功耗。
本申请实施例还公开了一种整机箱功耗的分配装置，应用于CMC，参见图4所示，包括处理器11和存储器12；其中，所述处理11执行所述存储器12中保存的计算机程序时实现以下步骤：
根据整机箱额定功耗和各节点的预设功耗，确定备用功耗总量；
向每个所述节点分配对应的预设功耗；
按照检测频率检测每个所述节点的实际功耗；
根据每个所述节点的实际功耗的变化速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率，对所述备用功耗总量与该节点的预设功耗进行再分配。
本发明实施例应用于整机箱额定功耗有限制的情况下，由CMC自动调节各节点的预设功耗，根据检测到的每个节点的实际功耗的变化和功耗利用率，重新调整该节点的预设功耗，从而保证节点在业务繁忙时能够有较大功耗支撑，以较优的性能处理业务，在业务较少时释放空闲功耗到备用功耗总量中，避免过多功耗占用，导致其他节点的处理效率受到功耗限制，最大限度提高了整机箱的有效利用率，降低了功耗空置，运行成本降低。
在一些具体的实施例中，每个所述节点对应的预设功耗，具体为根据该节点的节点等级确定的预设功耗。
在一些具体的实施例中，所述处理器11执行所述存储器12中保存的计算机子程序时，具体可以实现以下步骤：
当该节点的实际功耗的增加速度超出第一速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率超出第一利用率，则根据所述增加速度确定第一分配功耗，将所述第一分配功耗从所述备用功耗总量中分配至该节点的预设功耗中；
当该节点的实际功耗的降低速度超出第二速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率低于第二利用率，则根据所述降低速度确定第二分配功耗，将所述第二分配功耗从该节点的预设功耗中分配至所述备用功耗总量中。
在一些具体的实施例中，所述第一利用率不小于所述第二利用率。
在一些具体的实施例中，所述节点的实际功耗对应的变化速度具体为功耗利用率的变化速度。
在一些具体的实施例中，所述处理器11执行所述存储器12中保存的计算机子程序时，具体还可以实现以下步骤：
当所述备用功耗总量为零，确认符合功耗回收条件的所述节点为功耗待回收节点，对所述功耗待回收节点的预设功耗和所述备用功耗总量进行再分配。
在一些具体的实施例中，所述功耗回收条件具体为所述节点的功耗利用率低于第三利用率；
相应的，对任一所述功耗待回收节点的预设功耗和所述备用功耗总量进行再分配的过程，具体包括：
确定第三分配功耗，将所述第三分配功耗从该功耗待回收节点的预设功耗中分配至所述备用功耗总量中，同时使该功耗待回收节点的功耗利用率达到或超过第四利用率。
在一些具体的实施例中，所述处理器11执行所述存储器12中保存的计算机子程序时，具体还可以实现以下步骤：
当所述备用功耗总量为负，对所有所述节点中预设功耗和对应的实际功耗分别求差，得到剩余功耗，对所有所述剩余功耗求和，得到剩余功耗总量；
当所述备用功耗总量与所述剩余功耗总量相加为正，则按照所述备用功耗总量与所述剩余功耗总量的绝对值比，将每个所述节点的剩余功耗分配至所述备用功耗总量中；
当所述备用功耗总量与所述剩余功耗总量相加为负，则按照所有所述节点的实际功耗的比例，将所述整机箱额定功耗分配至每个节点的预设功耗。
进一步的，本实施例中的整机箱功耗的分配装置，还可以包括：
输入接口13，用于获取外界导入的计算机程序，并将获取到的计算机程序保存至所述存储器12中，还可以用于获取外界终端设备传输的各种指令和参数例如预设功耗和检测频率，获取每个节点的BMC传输的实际功耗，并传输至处理器11中，以便处理器11利用上述各种指令和参数展开相应的处理。本实施例中，所述输入接口13具体可以包括但不限于USB接口、串行接口、语音输入接口LAN接口、I2C接口、硬盘读取接口等。
输出接口14，用于将处理器11产生的各种数据输出至与其相连的终端 设备，以便于与输出接口14相连的其他终端设备能够获取到处理器11产生的各种数据，例如将再分配的预设功耗信息发送给对应的各节点。本实施例中，所述输出接口14具体可以包括但不限于LAN接口、I2C接口、USB接口、串行接口等。
通讯单元15，用于在分配装置和外部装置如服务器或客户端之间建立远程通讯连接，以便于该外部装置查看分配装置内部的状态信息、向分配装置发送参数指令。本实施例中，通讯单元15具体可以包括但不限于基于无线通讯技术或有线通讯技术的远程通讯单元。
键盘16，用于获取用户通过实时敲击键帽而输入的各种参数数据或指令。
显示器17，用于对功耗分配过程的相关信息进行实时显示，以便于用户及时地了解当前整机箱功耗的分配情况。
鼠标18，可以用于协助用户输入数据并简化用户的操作。
进一步的，本申请实施例还公开了一种计算机可读存储介质，这里所说的计算机可读存储介质包括随机存储器(RAM)、内存、只读存储器(ROM)、电可编程ROM、电可擦除可编程ROM、寄存器、硬盘、可移动硬盘、CD-ROM或技术领域内所公知的任意其他形式的存储介质。计算机可读存储介质中存储有计算机程序，所述计算机程序被处理器执行时实现以下步骤：
根据整机箱额定功耗和各节点的预设功耗，确定备用功耗总量；
向每个所述节点分配对应的预设功耗；
按照检测频率检测每个所述节点的实际功耗；
根据每个所述节点的实际功耗的变化速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率，对所述备用功耗总量与该节点的预设功耗进行再分配。
本发明实施例应用于整机箱额定功耗有限制的情况下，由CMC自动调节各节点的预设功耗，根据检测到的每个节点的实际功耗的变化和功耗利用率，重新调整该节点的预设功耗，从而保证节点在业务繁忙时能够有较 大功耗支撑，以较优的性能处理业务，在业务较少时释放空闲功耗到备用功耗总量中，避免过多功耗占用，导致其他节点的处理效率受到功耗限制，最大限度提高了整机箱的有效利用率，降低了功耗空置，运行成本降低。
在一些具体的实施例中，每个所述节点对应的预设功耗，具体为根据该节点的节点等级确定的预设功耗。
在一些具体的实施例中，所述计算机可读存储介质中存储的计算机子程序被处理器执行时，具体可以实现以下步骤：
当该节点的实际功耗的增加速度超出第一速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率超出第一利用率，则根据所述增加速度确定第一分配功耗，将所述第一分配功耗从所述备用功耗总量中分配至该节点的预设功耗中；
当该节点的实际功耗的降低速度超出第二速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率低于第二利用率，则根据所述降低速度确定第二分配功耗，将所述第二分配功耗从该节点的预设功耗中分配至所述备用功耗总量中。
在一些具体的实施例中，所述第一利用率不小于所述第二利用率。
在一些具体的实施例中，所述节点的实际功耗对应的变化速度具体为功耗利用率的变化速度。
在一些具体的实施例中，所述计算机可读存储介质中存储的计算机子程序被处理器执行时，具体可以实现以下步骤：
当所述备用功耗总量为零，确认符合功耗回收条件的所述节点为功耗待回收节点，对所述功耗待回收节点的预设功耗和所述备用功耗总量进行再分配。
在一些具体的实施例中，所述功耗回收条件具体为所述节点的功耗利用率低于第三利用率；
相应的，对任一所述功耗待回收节点的预设功耗和所述备用功耗总量进行再分配的过程，具体包括：
确定第三分配功耗，将所述第三分配功耗从该功耗待回收节点的预设功耗中分配至所述备用功耗总量中，同时使该功耗待回收节点的功耗利用 率达到或超过第四利用率。
在一些具体的实施例中，所述计算机可读存储介质中存储的计算机子程序被处理器执行时，具体可以实现以下步骤：
当所述备用功耗总量为负，对所有所述节点中预设功耗和对应的实际功耗分别求差，得到剩余功耗，对所有所述剩余功耗求和，得到剩余功耗总量；
当所述备用功耗总量与所述剩余功耗总量相加为正，则按照所述备用功耗总量与所述剩余功耗总量的绝对值比，将每个所述节点的剩余功耗分配至所述备用功耗总量中；
当所述备用功耗总量与所述剩余功耗总量相加为负，则按照所有所述节点的实际功耗的比例，将所述整机箱额定功耗分配至每个节点的预设功耗。
需要说明的是，本发明中所述的参数a、b、c、d、e、f、g、j、k、m、p、q仅为阐述方便所做的区分，不同字母所代表的参数可以相同或不同。
本发明中所述的角度α、β、λ、θ、γ、δ、η亦为阐述方便所做的区分，符号本身不具有任何意义。
需要说明的是，对于前述的各方法实施例，为了简单描述，故将其都表述为一系列动作的组合，但本领域的技术人员应该知悉，本发明并不受所描述的动作的顺序的限制，因为依据本发明，某些步骤可以采用其它顺序或者同时进行。其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作和模块并不一定是本发明所必须的。
最后，还需要说明的是，本领域技术人员可以理解上述实施例的各种方法中的全部或者部分步骤是可以通过程序来指令相关的硬件来完成的，该程序可以存储于一计算机可读存储单元中。本发明所述的所有实施例中所述的存储单元包括：只读存储器、随机存储器、磁盘或等等。
在本文中，诸如术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备 不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
本说明书中各个实施例采用递进的方式描述，每个实施例重点说明的都是与其他实施例的不同之处，各个实施例之间相同相似部分互相参见即可。
对所公开的实施例的上述说明，使本领域专业技术人员能够实现或使用本发明。对这些实施例的多种修改对本领域的专业技术人员来说将是显而易见的，本文中所定义的一般原理可以在不脱离本发明的精神或范围的情况下，在其它实施例中实现。因此，本发明将不会被限制于本文所示的这些实施例，而是要符合与本文所公开的原理和新颖特点相一致的最宽的范围。

Claims (11)

  1. 一种整机箱功耗的分配方法，其特征在于，应用于CMC，包括：
    根据整机箱额定功耗和各节点的预设功耗，确定备用功耗总量；
    向每个所述节点分配对应的预设功耗；
    按照检测频率检测每个所述节点的实际功耗；
    根据每个所述节点的实际功耗的变化速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率，对所述备用功耗总量与该节点的预设功耗进行再分配。
  2. 根据权利要求1所述分配方法，其特征在于，每个所述节点对应的预设功耗，具体为根据该节点的节点等级确定的预设功耗。
  3. 根据权利要求1所述分配方法，其特征在于，根据任一所述节点的实际功耗的变化速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率，对所述备用功耗总量与该节点的预设功耗进行再分配的过程，具体包括：
    当该节点的实际功耗的增加速度超出第一速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率超出第一利用率，则根据所述增加速度确定第一分配功耗，将所述第一分配功耗从所述备用功耗总量中分配至该节点的预设功耗中；
    当该节点的实际功耗的降低速度超出第二速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率低于第二利用率，则根据所述降低速度确定第二分配功耗，将所述第二分配功耗从该节点的预设功耗中分配至所述备用功耗总量中。
  4. 根据权利要求3所述分配方法，其特征在于，
    所述第一利用率不小于所述第二利用率。
  5. 根据权利要求4所述分配方法，其特征在于，所述节点的实际功耗对应的变化速度具体为功耗利用率的变化速度。
  6. 根据权利要求1至5任一项所述分配方法，其特征在于，还包括：
    当所述备用功耗总量为零，确认符合功耗回收条件的所述节点为功耗待回收节点，对所述功耗待回收节点的预设功耗和所述备用功耗总量进行 再分配。
  7. 根据权利要求6所述分配方法，其特征在于，
    所述功耗回收条件具体为所述节点的功耗利用率低于第三利用率；
    相应的，对任一所述功耗待回收节点的预设功耗和所述备用功耗总量进行再分配的过程，具体包括：
    确定第三分配功耗，将所述第三分配功耗从该功耗待回收节点的预设功耗中分配至所述备用功耗总量中，同时使该功耗待回收节点的功耗利用率达到或超过第四利用率。
  8. 根据权利要求7所述分配方法，其特征在于，还包括：
    当所述备用功耗总量为负，对所有所述节点中预设功耗和对应的实际功耗分别求差，得到剩余功耗，对所有所述剩余功耗求和，得到剩余功耗总量；
    当所述备用功耗总量与所述剩余功耗总量相加为正，则按照所述备用功耗总量与所述剩余功耗总量的绝对值比，将每个所述节点的剩余功耗分配至所述备用功耗总量中；
    当所述备用功耗总量与所述剩余功耗总量相加为负，则按照所有所述节点的实际功耗的比例，将所述整机箱额定功耗分配至每个所述节点的预设功耗。
  9. 一种整机箱功耗的分配系统，其特征在于，应用于CMC，包括：
    备用确定模块，用于根据整机箱额定功耗和各节点的预设功耗，确定备用功耗总量；
    第一分配模块，用于向每个所述节点分配对应的预设功耗；
    功耗检测模块，用于按照检测频率检测每个所述节点的实际功耗；
    第二分配模块，用于根据每个所述节点的实际功耗的变化速度和/或该节点的实际功耗占该节点的预设功耗的功耗利用率，对所述备用功耗总量与该节点的预设功耗进行再分配。
  10. 一种整机箱功耗的分配装置，其特征在于，应用于CMC，包括：
    存储器，用于存储计算机程序；
    处理器，用于执行所述计算机程序时实现如权利要求1至8任一项所 述整机箱功耗的分配方法的步骤。
  11. 一种可读存储介质，其特征在于，应用于CMC，所述可读存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现如权利要求1至8任一项所述整机箱功耗的分配方法的步骤。
PCT/CN2019/108980 2019-06-27 2019-09-29 一种整机箱功耗的分配方法、系统、装置及可读存储介质 WO2020258569A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/281,590 US11586264B2 (en) 2019-06-27 2019-09-29 Allocation method, system and device for power consumption of complete machine box, and readable storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910568193.3 2019-06-27
CN201910568193.3A CN110399216B (zh) 2019-06-27 2019-06-27 一种整机箱功耗的分配方法、系统、装置及可读存储介质

Publications (1)

Publication Number Publication Date
WO2020258569A1 true WO2020258569A1 (zh) 2020-12-30

Family

ID=68324282

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/108980 WO2020258569A1 (zh) 2019-06-27 2019-09-29 一种整机箱功耗的分配方法、系统、装置及可读存储介质

Country Status (3)

Country Link
US (1) US11586264B2 (zh)
CN (1) CN110399216B (zh)
WO (1) WO2020258569A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111414070B (zh) * 2020-03-06 2021-10-15 苏州浪潮智能科技有限公司 一种机箱功耗管理方法、系统及电子设备和存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060177192A1 (en) * 2005-02-09 2006-08-10 Cisco Technology, Inc. Method and apparatus for negotiating power between power sourcing equipment and powerable devices
CN102208986A (zh) * 2011-03-22 2011-10-05 曙光信息产业(北京)有限公司 一种集群功耗分配与控制方法
CN102497275A (zh) * 2011-12-02 2012-06-13 曙光信息产业(北京)有限公司 一种基于功耗池的集群功耗分配方法
CN103389791A (zh) * 2013-06-25 2013-11-13 华为技术有限公司 数据系统的功率控制方法及装置
CN108874112A (zh) * 2018-06-28 2018-11-23 郑州云海信息技术有限公司 一种整机柜功耗限制方法、装置、设备及存储介质

Family Cites Families (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8375228B2 (en) * 2008-08-06 2013-02-12 International Business Machines Corporation Multiple-node system power utilization management
US9047083B2 (en) * 2008-09-15 2015-06-02 Vmware, Inc. Reducing power consumption in a server cluster
WO2010032501A1 (ja) * 2008-09-17 2010-03-25 株式会社日立製作所 情報処理システムの運用管理方法
CN102419718A (zh) * 2011-10-28 2012-04-18 浪潮(北京)电子信息产业有限公司 资源调度方法
US9116181B2 (en) * 2011-12-29 2015-08-25 Huawei Technologies Co., Ltd. Method, apparatus, and system for virtual cluster integration
US9201486B2 (en) * 2012-02-23 2015-12-01 Cisco Technology, Inc. Large scale dynamic power budget adjustments for optimizing power utilization in a data center
US8843772B2 (en) * 2012-05-22 2014-09-23 Dell Products Lp Systems and methods for dynamic power allocation in an information handling system environment
US10067547B2 (en) * 2012-06-28 2018-09-04 Intel Corporation Power management control of remote servers
US9110972B2 (en) * 2012-11-07 2015-08-18 Dell Products L.P. Power over ethernet dynamic power allocation system
US9529642B2 (en) * 2013-03-28 2016-12-27 Vmware, Inc. Power budget allocation in a cluster infrastructure
US9753520B2 (en) * 2013-12-23 2017-09-05 Dell Products, L.P. Predictive power capping and power allocation to computing nodes in a rack-based information handling system
US10102035B2 (en) * 2014-02-27 2018-10-16 Intel Corporation Techniques for computing resource discovery and management in a data center
US9454200B2 (en) * 2014-05-05 2016-09-27 International Business Machines Corporation Budgeting for power consumption in a chassis environment that includes a plurality of integrated technology elements
US10429909B2 (en) * 2015-06-01 2019-10-01 Hewlett Packard Enterprise Development Lp Managing power in a high performance computing system for resiliency and cooling
US9983652B2 (en) * 2015-12-04 2018-05-29 Advanced Micro Devices, Inc. Balancing computation and communication power in power constrained clusters
US9880599B1 (en) * 2016-10-12 2018-01-30 International Business Machines Corporation Priority-aware power capping for hierarchical power distribution networks
JP6799255B2 (ja) * 2016-11-22 2020-12-16 富士通株式会社 ジョブ消費電力推定プログラム、並列処理装置およびジョブ消費電力推定方法
US10423217B1 (en) * 2017-07-14 2019-09-24 Cisco Technology, Inc. Dynamic power capping of multi-server nodes in a chassis based on real-time resource utilization
CN109388488B (zh) * 2017-08-02 2022-11-25 联想企业解决方案(新加坡)有限公司 计算机系统中的功率分配
CN107908479B (zh) * 2017-12-11 2021-03-02 北京奇艺世纪科技有限公司 一种节点资源分配方法及装置
US10824215B2 (en) * 2018-07-31 2020-11-03 Nutanix, Inc. Managing power budget of multiple computing node clusters in a computing rack system
US11347288B2 (en) * 2020-10-13 2022-05-31 Hewlett Packard Enterprise Development Lp Power management in a blade enclosure

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060177192A1 (en) * 2005-02-09 2006-08-10 Cisco Technology, Inc. Method and apparatus for negotiating power between power sourcing equipment and powerable devices
CN102208986A (zh) * 2011-03-22 2011-10-05 曙光信息产业(北京)有限公司 一种集群功耗分配与控制方法
CN102497275A (zh) * 2011-12-02 2012-06-13 曙光信息产业(北京)有限公司 一种基于功耗池的集群功耗分配方法
CN103389791A (zh) * 2013-06-25 2013-11-13 华为技术有限公司 数据系统的功率控制方法及装置
CN108874112A (zh) * 2018-06-28 2018-11-23 郑州云海信息技术有限公司 一种整机柜功耗限制方法、装置、设备及存储介质

Also Published As

Publication number Publication date
US20210365095A1 (en) 2021-11-25
US11586264B2 (en) 2023-02-21
CN110399216A (zh) 2019-11-01
CN110399216B (zh) 2021-10-15

Similar Documents

Publication Publication Date Title
EP3507692B1 (en) Resource oversubscription based on utilization patterns in computing systems
US8473768B2 (en) Power control apparatus and method for cluster system
US20110161696A1 (en) Reducing energy consumption in a cloud computing environment
WO2018040360A1 (zh) 一种整机柜功耗限制方法及装置
CN103188277A (zh) 负载能耗管理系统、方法和服务器
WO2013097396A1 (zh) 一种虚拟化集群整合方法、装置及系统
CN113535409B (zh) 一种面向能耗优化的无服务器计算资源分配系统
CN103986766A (zh) 自适应负载均衡作业任务调度方法及装置
CN111625080B (zh) 一种服务器节能方法、装置及电子设备和存储介质
CN102208986B (zh) 一种集群功耗分配与控制方法
TW201324357A (zh) 虛擬機叢集之綠能管理方法
CN103095846A (zh) 云计算资源的用户个性化调度方法和系统
Zhang et al. A new energy efficient VM scheduling algorithm for cloud computing based on dynamic programming
Song et al. Server consolidation energy-saving algorithm based on resource reservation and resource allocation strategy
CN114048006A (zh) 虚拟机动态迁移方法、装置以及存储介质
WO2020258569A1 (zh) 一种整机箱功耗的分配方法、系统、装置及可读存储介质
CN103279184A (zh) 一种基于数据挖掘的云服务器节能系统
CN108574600B (zh) 云计算服务器的功耗和资源竞争协同控制的服务质量保障方法
CN109600977B (zh) 一种散热控制方法及电子设备
CN104796673A (zh) 一种面向能耗优化的云视频监控系统任务接入方法
CN108073449B (zh) 一种虚拟机动态放置方法
CN110308991B (zh) 一种基于随机任务的数据中心节能优化方法及系统
Carrega et al. Energy-aware consolidation scheme for data center cloud applications
WO2018009103A1 (en) Power manager and method performed thereby for managing power of a datacentre
Arabas Modeling and simulation of hierarchical task allocation system for energy-aware HPC clouds

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19935012

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19935012

Country of ref document: EP

Kind code of ref document: A1