CN115826729A - Hard disk power management method, device, electronic equipment, medium and product - Google Patents
Hard disk power management method, device, electronic equipment, medium and product Download PDFInfo
- Publication number
- CN115826729A CN115826729A CN202211261792.9A CN202211261792A CN115826729A CN 115826729 A CN115826729 A CN 115826729A CN 202211261792 A CN202211261792 A CN 202211261792A CN 115826729 A CN115826729 A CN 115826729A
- Authority
- CN
- China
- Prior art keywords
- hard disk
- power consumption
- power supply
- current
- target power
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
- 238000007726 management method Methods 0.000 title claims abstract description 119
- 238000000034 method Methods 0.000 claims description 36
- 238000003066 decision tree Methods 0.000 claims description 26
- 238000004590 computer program Methods 0.000 claims description 24
- 238000012549 training Methods 0.000 claims description 17
- 230000001419 dependent effect Effects 0.000 claims description 11
- 238000003860 storage Methods 0.000 claims description 7
- 238000002372 labelling Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 abstract description 5
- 238000005265 energy consumption Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 14
- 230000008569 process Effects 0.000 description 8
- 238000007405 data analysis Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 230000004907 flux Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 230000017525 heat dissipation Effects 0.000 description 3
- 230000003993 interaction Effects 0.000 description 3
- 238000002360 preparation method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- 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
Landscapes
- Power Sources (AREA)
Abstract
The embodiment of the invention provides a hard disk power supply management method, a hard disk power supply management device, electronic equipment, a medium and a product. In the embodiment of the invention, the BMC determines the optimal power consumption value corresponding to the current client scene and the corresponding hard disk working mode according to the current client running scene and the pre-trained target power consumption prediction model, determines the corresponding target power supply parameter according to the power consumption value and the working mode, writes the target power supply parameter into the main control chip of the SSD, and realizes the dynamic management of the hard disk power supply state, thereby realizing the dynamic balance of the service target and the energy consumption. In the embodiment of the invention, the target power consumption prediction model learns the corresponding relation among the client operation scene, the hard disk operation mode and the target power consumption value, so that the target power consumption prediction model can determine the corresponding operation mode according to the client operation scene, and establish a mapping relation with the corresponding target power consumption value to realize the purpose of applying the specific target power consumption value to the specific operation scene.
Description
Technical Field
The embodiment of the invention relates to the field of data processing, in particular to a hard disk power management method, a hard disk power management device, electronic equipment, a medium and a product.
Background
With the maturity of Solid State Disk (SSD) technology, the SSD has gradually replaced the traditional mechanical hard Disk on an enterprise-level server with high performance throughput, and is widely applied to various fields such as industrial control, video monitoring, network terminal, navigation device, and the like. The storage medium of the SSD hard disk is different from the mechanical hard disk, and the cost per unit area is high, and if the power consumption is too high, the heat dissipation efficiency is affected and the problem of speed drop is caused, which causes a cost loss. The requirements for the SSD are different in different application scenes after a client purchases a server, at present, due to the fact that the application scenes of the client are more, and the requirements for the SSD in different application scenes are different, if the power state of the SSD cannot be changed in real time, the SSD is easily overhigh in power consumption and poor in heat dissipation, the flash memory is damaged finally, the operation cost is increased, or the power consumption is low, and the requirement for sudden data access cannot be met, so that the service performance is reduced.
Therefore, a hard disk power management method is needed to dynamically manage the power status of the hard disk.
Disclosure of Invention
The embodiment of the invention provides a hard disk power supply management method, a hard disk power supply management device, electronic equipment, a medium and a product, which are used for dynamically managing the power supply state of a hard disk.
The first aspect of the embodiments of the present invention provides a hard disk power management method, which is applied to BMC, and the method includes:
acquiring current data read-write delay parameters and determining a current client operation scene;
determining a current hard disk working mode and a current target power consumption value according to the current client running scene and a pre-trained target power consumption prediction model, wherein the target power consumption prediction model learns the corresponding relation among the client running scene, the hard disk working mode and the target power consumption value;
determining a current target power supply parameter according to the current hard disk working mode and the current target power consumption value;
and writing the current target power supply parameter into a main control chip of a hard disk.
Optionally, the method further comprises:
obtaining a sample client operation scene, labeling the sample client operation scene, and obtaining a target power consumption value label corresponding to the sample client operation scene;
performing model training on the read-write delay characteristic, the hard disk working mode characteristic and the target power consumption value label of the sample client operation scene by adopting a decision tree algorithm to obtain a target power consumption prediction model;
the working mode is characterized by the proportion of two access modes, namely sequential read-write access and random read-write access.
Optionally, the method further comprises:
acquiring historical hard disk power supply parameters, corresponding historical power consumption values and historical working modes through a power supply management interface provided by an NVEM (network video and Environment) protocol;
taking the historical hard disk power supply parameters as independent variables, and taking corresponding historical power consumption values and historical working modes as dependent variables, and determining the corresponding relation among the hard disk power supply parameters, the power consumption values and the working modes;
determining a current target power supply parameter according to the current hard disk working mode and the current target power consumption value, wherein the determining comprises the following steps:
and determining the current target power supply parameter according to the current hard disk working mode, the current target power consumption value and the corresponding relation between the hard disk power supply parameter, the power consumption value and the working mode.
Optionally, after writing the current target power parameter into the main control chip of the SSD, the method further includes:
and acquiring the current actual power supply parameter through a power supply management interface provided by an NVME protocol so as to realize the closed loop of the hard disk power supply management.
A second aspect of an embodiment of the present invention provides a hard disk power management method, which is applied to a main control chip of a hard disk, and the method includes:
receiving a current target power supply parameter written by the BMC through a power supply management interface provided by an NVME protocol;
managing the power state of the hard disk according to the current target power parameter;
the BMC is used for acquiring current data read-write delay parameters and determining a current client operation scene; determining a current hard disk working mode and a current target power consumption value according to the current client running scene and a pre-trained target power consumption prediction model, wherein the target power consumption prediction model learns the corresponding relation among the client running scene, the hard disk working mode and the target power consumption value; and determining the current target power supply parameter according to the current hard disk working mode and the current target power consumption value.
Optionally, after the hard disk power state is managed according to the current target power parameter, the method further includes:
acquiring current actual power supply parameters;
and feeding back the current actual power supply parameter to the BMC so that the BMC can acquire the current actual power supply parameter and realize the closed loop of hard disk power supply management.
Optionally, the method further comprises:
sending historical hard disk power supply parameters, corresponding historical power consumption values and historical working modes to the BMC through a power supply management interface provided by an NVEM (network video and Environment) protocol;
the BMC is further used for acquiring historical hard disk power supply parameters, corresponding historical power consumption values and historical working modes through a power supply management interface provided by the NVEM protocol, taking the historical hard disk power supply parameters as independent variables, taking the corresponding historical power consumption values and historical working modes as dependent variables, and determining the corresponding relation among the hard disk power supply parameters, the power consumption values and the working modes.
A third aspect of an embodiment of the present invention provides a hard disk power management device, which is applied to a BMC, where the device includes:
the first determining module is used for acquiring the current data read-write delay parameter and determining the current client operation scene;
the second determining module is used for determining a current hard disk working mode and a current target power consumption value according to the current client running scene and a pre-trained target power consumption prediction model, and the target power consumption prediction model learns the corresponding relation among the client running scene, the hard disk working mode and the target power consumption value;
the third determining module is used for determining the current target power supply parameter according to the current hard disk working mode and the current target power consumption value;
and the writing module is used for writing the current target power supply parameter into a main control chip of a hard disk.
Optionally, the apparatus further comprises:
the system comprises a marking module, a power consumption value obtaining module and a power consumption value setting module, wherein the marking module is used for obtaining a sample client operation scene, marking the sample client operation scene and obtaining a target power consumption value label corresponding to the sample client operation scene;
the training module is used for performing model training on the read-write delay characteristic, the hard disk working mode characteristic and the target power consumption value label of the sample client operation scene by adopting a decision tree algorithm to obtain a target power consumption prediction model;
the working mode is characterized by the proportion of two access modes, namely sequential read-write access and random read-write access.
Optionally, the apparatus further comprises:
the acquisition module is used for acquiring historical hard disk power supply parameters, corresponding historical power consumption values and historical working modes through a power supply management interface provided by an NVEM protocol;
the fourth determining module is used for determining the corresponding relation among the power supply parameters, the power consumption values and the working modes of the hard disk by taking the historical power supply parameters as independent variables and the corresponding historical power consumption values and the historical working modes as dependent variables;
the third determining module is specifically configured to:
and determining the current target power supply parameter according to the current hard disk working mode, the current target power consumption value and the corresponding relation between the hard disk power supply parameter, the power consumption value and the working mode.
Optionally, the apparatus further comprises:
and the acquisition module is used for acquiring the current actual power supply parameters through a power supply management interface provided by the NVME protocol so as to realize the closed loop of the hard disk power supply management.
A fourth aspect of the embodiments of the present invention provides a hard disk power management device, which is applied to a main control chip of a hard disk, and includes:
the receiving module is used for receiving the current target power supply parameter written by the BMC through a power supply management interface provided by the NVME protocol;
the management module is used for managing the power state of the hard disk according to the current target power parameter;
the BMC is used for acquiring current data read-write delay parameters and determining a current client operation scene; determining a current hard disk working mode and a current target power consumption value according to the current client running scene and a pre-trained target power consumption prediction model, wherein the target power consumption prediction model learns the corresponding relation among the client running scene, the hard disk working mode and the target power consumption value; and determining the current target power supply parameter according to the current hard disk working mode and the current target power consumption value.
Optionally, the apparatus further comprises:
the acquisition module is used for acquiring the current actual power supply parameter;
and the feedback module is used for feeding the current actual power supply parameter back to the BMC so that the BMC can acquire the current actual power supply parameter and realize the closed loop of hard disk power supply management.
Optionally, the apparatus further comprises:
the sending module is used for sending the historical hard disk power supply parameters, the corresponding historical power consumption values and the historical working modes to the BMC through a power supply management interface provided by an NVEM protocol;
the BMC is further used for acquiring historical hard disk power supply parameters, corresponding historical power consumption values and historical working modes through a power supply management interface provided by the NVEM protocol, taking the historical hard disk power supply parameters as independent variables, taking the corresponding historical power consumption values and the corresponding historical working modes as dependent variables, and determining the corresponding relation among the hard disk power supply parameters, the power consumption values and the working modes.
A fifth aspect of the embodiments of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the hard disk power management method according to the first aspect or the second aspect of the present invention when executing the computer program.
A sixth aspect of the embodiments of the present invention provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of the hard disk power management method according to the first or second aspect of the present invention.
A seventh aspect of the embodiments of the present invention provides a computer program product, which includes computer programs/instructions, and when executed by a processor, the computer programs/instructions implement the steps of the hard disk power management method according to the first aspect or the second aspect of the present invention.
By adopting the hard disk power management method provided by the embodiment of the invention, the BMC determines the optimal power consumption value corresponding to the current client scene and the corresponding hard disk working mode according to the current client running scene and the pre-trained target power consumption prediction model, determines the corresponding target power supply parameter according to the power consumption value and the working mode, and writes the target power supply parameter into the main control chip of the SSD to realize the dynamic management of the hard disk power supply state, thereby realizing the dynamic balance of the service target and the energy consumption.
In the embodiment of the invention, the target power consumption prediction model learns the corresponding relation among the client operation scene, the hard disk operation mode and the target power consumption value, so that the target power consumption prediction model can determine the corresponding operation mode according to the client operation scene, and establish a mapping relation with the corresponding target power consumption value to realize the purpose of applying the specific target power consumption value to the specific operation scene.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive labor.
FIG. 1 is a flow chart of a hard disk power management method according to an embodiment of the present invention;
FIG. 2 is a flow chart of a hard disk power management method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating an interaction flow of a hard disk power management method according to an embodiment of the present invention;
FIG. 4 is a flowchart of a hard disk power management method according to an embodiment of the present invention;
FIG. 5 is a flowchart of a hard disk power management method according to an embodiment of the present invention;
FIG. 6 is a block diagram of a hard disk power management device according to an embodiment of the present invention;
fig. 7 is a block diagram of a hard disk power management apparatus according to an embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
In the related technical solution, no matter what the server system has the requirement on the performance of the hard disk, the hard disk always works in the same power state, and even when the server system does not have the requirement on the read-write performance of the hard disk, the hard disk is also in the power state with high power, so that the system cannot save energy according to the performance requirement of actual use.
Based on this, the embodiments of the present invention provide the inventive concept: and dynamically managing the power state of the hard disk according to the actual requirement of the server system on the performance of the hard disk.
Specifically, referring to fig. 1, a flowchart of a hard disk power management method according to an embodiment of the present invention is shown, where the hard disk power management method according to the embodiment of the present invention is applied to a BMC (baseboard management controller), where the BMC is an independent processor embedded in a server motherboard and passes through an IPMB and an LPC
The interface collection of (low-pin-count-interface), SMBus, etc. communicates with other software and hardware components in the host computer, and provides inquiry and control functions to the local host computer/remote server through the interfaces of network, serial/modem, PCI, etc.
Specifically, the hard disk power management method provided by the embodiment of the present invention may include the following steps:
s101, collecting current data read-write delay parameters and determining a current client operation scene.
In the embodiment of the present invention, the current data read-write delay parameter includes: a read delay parameter and a write delay parameter. In practical application, different client operation scenes correspond to different data read-write delay parameters. The client operation scene refers to a main service executed by the server after the client purchases the server, namely, the operation scene of the server, and correspondingly, the operation scene of the hard disk, and the client operation scene is divided into videos, various databases, financial transactions, internet of things and the like. Due to the fact that the data reading and writing requirements are different under different operation scenes, correspondingly, the data reading and writing delay parameters are different.
In the embodiment of the invention, when the server runs, the BMC can acquire the current data read-write delay parameter from the server, further determine the current client running scene, and further determine the target power consumption and the hard disk working mode corresponding to the current client running scene.
And S102, determining a current hard disk working mode and a current target power consumption value according to the current client running scene and a pre-trained target power consumption prediction model.
The target power consumption prediction model learns the corresponding relation among a client operation scene, a hard disk working mode and a target power consumption value.
In the embodiment of the invention, parameters (such as data read-write delay parameters, hard disk power consumption, hard disk working modes and the like) in different client running scenes can be captured, so that a plurality of client running scenes and a plurality of corresponding parameters can be used as training samples, a preset model is trained, and a target power consumption prediction model is obtained, so that the target power consumption prediction model can determine the optimal power consumption value and the hard disk working mode matched with the client running scenes according to the client running scenes identified by the BMC.
The preset model can be a machine learning model based on a decision tree algorithm, and can also be other deep learning models applied to maturity in the related technology.
In the embodiment of the invention, the current hard disk working mode and the current target power consumption value are the optimal hard disk working mode and the optimal power consumption value which are matched with the current client running scene.
S103, determining current target power supply parameters according to the current hard disk working mode and the current target power consumption value.
In the embodiment of the invention, the BMC can also determine the corresponding target power supply parameter according to the working mode and the power consumption value of the SSD. The target Power supply parameter may be a maximum Power, an entry State time, an exit State time, a relative read flux, a relative read delay, a relative write flux, a relative write delay, and the like in a Power State Descriptor structure of a hard disk provided by a Non Volatile Memory access Express (NVEM) protocol.
Specifically, in the embodiment of the present invention, the BMC may obtain the historical power supply parameters of the SSD, and the corresponding operating mode and power consumption value, thereby determining the corresponding relationship between the SSD power supply parameters and the power consumption value, and establishing the mapping relationship, thereby determining the current target power supply parameter of the SSD according to the mapping relationship and the current hard disk operating mode and the current target power consumption value determined in step S102.
And S104, writing the current target power supply parameter into a main control chip of the hard disk.
In the embodiment of the invention, the BMC can write the current target power supply parameter into the main control chip of the hard disk through the power supply management interface provided by the NVEM protocol so that the main control chip controls the SSD power supply state and realizes the dynamic management of the SSD power supply state.
Different power supply parameters (maximum power, entering state time, exiting state time, relative read flux, relative read delay, relative write flux, relative write delay parameters) correspond to different hard disk working power supply states, and the power supply parameters can be specifically written into the SSD master control chip and stored into the state query table.
In practical application, when the BMC operates the hard disk power management method provided by the embodiment of the invention, the BMC may determine the requirements of the current client operation scene on the hard disk according to the data read-write delay parameter acquired in real time, determine the optimal target power consumption value and working mode of the hard disk according to the current operation scene, and control the power state of the hard disk. When the data read-write delay parameter collected in real time changes, the BMC can re-determine the client operation scene, and further re-determine the optimal power consumption value and the optimal working mode of the hard disk, so as to dynamically control the power state of the hard disk along with the change of the client operation scene.
In the embodiment of the invention, the BMC can also control the power state of the hard disk to dynamically change at different time periods according to the change of data reading and writing requirements at different time periods in a client running scene. Further, in the embodiment of the present invention, for a certain server, the BMC may further collect changes of an operation scene of the server at different time periods, and further dynamically control the power state of the hard disk according to the different time periods.
For example, the current operation scene is a video scene, the data read-write delay parameter is high, and the hard disk is required to operate in a high-power consumption working mode, after the BMC determines the data read-write delay condition, the corresponding target power consumption value and working mode can be determined based on a pre-trained target power consumption prediction model, and the power supply state of the hard disk is controlled to operate in the high-power consumption working mode.
For another example, when the current operation scene is changed into a picture scene, the data read-write delay parameter is low, the hard disk can operate in a low-power consumption working mode, and after determining the data read-write delay condition, the BMC can determine a corresponding target power consumption value and a working mode based on a pre-trained target power consumption prediction model, and control the power state of the hard disk so that the hard disk operates in the low-power consumption working mode.
Therefore, in the embodiment of the invention, when the data reading and writing requirements of the server are high, the hard disk can be controlled to operate in the high-power-consumption working mode to adapt to high-demand data access, and when the data reading and writing requirements of the server are low, the hard disk can be controlled to operate in the low-power-consumption working mode to avoid the problems that the SSD is high in power consumption and poor in heat dissipation, and finally the flash memory is damaged to increase the operation cost.
Referring to fig. 2, a flowchart of another hard disk power management method according to an embodiment of the present invention is shown, where the another hard disk power management method according to the embodiment of the present invention is applied to a BMC. Specifically, the hard disk power management method provided by the embodiment of the present invention may include the following steps:
s201, obtaining a sample client operation scene, labeling the sample client operation scene, and obtaining a target power consumption value label corresponding to the sample client operation scene.
In the embodiment of the invention, the BMC can collect a plurality of client operation scenes and related parameters as sample client operation scenes, and labels the sample client operation scenes according to the target power consumption value corresponding to each sample client operation scene to obtain the labeled training samples.
S202, performing model training on the read-write delay characteristic, the hard disk working mode characteristic and the target power consumption value label of the sample client operation scene by adopting a decision tree algorithm to obtain the target power consumption prediction model.
A decision tree algorithm is a method of approximating discrete function values. It is a typical classification method that first processes the data, generates readable rules and decision trees using a generalised algorithm, and then uses the decisions to analyze the new data. In essence, a decision tree is a process of classifying data through a series of rules.
The basic idea of the decision tree algorithm mainly comprises:
1) The decision tree starts with a single node representing a training sample.
2) If the samples are all in the same class, the node becomes a tree node and is labeled with the class.
3) Otherwise, the algorithm selects the attribute with the most classification capability as the current node of the decision tree.
4) According to the difference of attribute values of the current decision node, a training sample data set is divided into a plurality of subsets, each value forms a branch, and a plurality of values form a plurality of branches. And repeating the previous steps aiming at the subset obtained in the previous step, and further forming a decision tree on each divided sample. Once an attribute appears on a node, it does not have to be considered on any descendant of the node.
5) The recursive partitioning step stops only when one of the following conditions holds:
(1) all samples of a given node belong to the same class.
(2) No remaining attributes are available to further divide the sample. In this case. Using majority voting, a given node is converted into a leaf node, and the class with the largest number of members in the sample is used as a class mark, and the class distribution of the node sample can be stored.
(3) If a branch does not satisfy a sample of existing classes in the branch, a tree node is created with the majority of the classes of samples.
The input of the decision tree algorithm is a group of samples with class labels, and the result of the algorithm operation is a binary tree or a multi-branch tree. The internal nodes (non-leaf nodes) of the binary tree are typically represented as a logical decision, e.g., of the form a = a j Wherein a is an attribute, a j Is all values of the attribute: the edges of the decision tree are the branching results of the logical decision. The internal nodes of the multi-branch tree are attributes, edges are all values of the attributes, several edges exist when several attribute values exist, and leaf nodes of the tree are labeled by categories.
In the embodiment of the invention, parameters in different client operation scenes can be captured, so that a decision tree structure with multiple layers and multiple feature points is formed. In the embodiment of the present invention, a service scene of a client may be divided into nodes 1 (for example, data read-write delay characteristics may be included), and each feature point in the nodes 1 may be continuously divided to form more detailed application scene characteristics (nodes 2). For the SSD hard disk, the node 2 is a ratio of the sequential read-write access to the random read-write access, i.e. a working mode.
In brief, in the embodiment of the present invention, the hard disk operating mode and the hard disk power consumption value corresponding to the client operating scenario may be determined through the operation of the decision tree algorithm. For example, in the embodiment of the present invention, the decision tree algorithm may obtain a mapping relationship between the client operation scenario and the corresponding optimal power consumption value based on the data read-write delay characteristic corresponding to the client operation scenario and the ratio of the sequential read-write access and the random read-write access of the corresponding SSD hard disk.
In the embodiment of the invention, the working mode is characterized by the proportion of two access modes of sequential read-write access and random read-write access. Specifically, the two access modes consume different powers in different proportions, and different data read and write delay performances are generated.
S203, acquiring historical hard disk power supply parameters, corresponding historical power consumption values and historical working modes through a power supply management interface provided by the NVEM protocol.
In the embodiment of the invention, the BMC can acquire the power parameters of the hard disk, the corresponding power consumption value and the working mode of the hard disk in real time through the power management interface provided by the NVEM protocol, and analyze the acquired data as historical data so as to determine the corresponding relation among the power parameters, the power consumption value and the working mode.
And S204, determining the corresponding relation among the power supply parameters, the power consumption values and the working modes of the hard disk by taking the historical power supply parameters as independent variables and the corresponding historical power consumption values and historical working modes as dependent variables.
In the embodiment of the present invention, the hard disk power supply parameter may be specifically used as an independent variable, the corresponding power consumption value and the corresponding operating mode may be used as a dependent variable, and the corresponding relationship between the hard disk power supply parameter and the power consumption value and the operating mode may be determined through data analysis of a preset duration.
In the embodiment of the invention, the corresponding relation between the hard disk power supply parameter and the power consumption value and the working mode can be displayed on a visual interface, so that technical personnel can observe and analyze the data. Specifically, the display may be performed in the form of a data analysis table, or may be performed in the form of a relationship curve.
Optionally, in this embodiment of the present invention, the steps S201 to S202 as the model training step may be executed simultaneously with the data analysis step provided in S203 to S204, or may be executed sequentially, specifically, the model training step (S201 to S202) may be executed first, and then the data analysis step (S203 to S204) may be executed, or the data analysis step (S203 to S204) may be executed first, and then the model training step (S201 to S202) may be executed, which is not limited in this embodiment of the present invention.
And S205, acquiring current data read-write delay parameters and determining the current client operation scene.
Similar to the aforementioned step S101, the description is omitted here.
And S206, determining the current hard disk working mode and the current target power consumption value according to the current client running scene and a pre-trained target power consumption prediction model.
In the embodiment of the invention, the optimal target power consumption value and the optimal working mode of the client operation scene and the hard disk can be accurately identified by using the decision tree algorithm in S201 and S202. In practical application, after a client starts a power state function of automatically switching an SSD, the BMC can collect data, run a decision tree algorithm, and finally accurately identify a client running scene, an optimal target power consumption value and a working mode of the SSD according to a decision tree result of long-term training.
And S207, determining the current target power supply parameter according to the current hard disk working mode, the current target power consumption value, the corresponding relation between the hard disk power supply parameter and the power consumption value as well as the working mode.
In the embodiment of the present invention, the BMC may determine the current target power supply parameter according to the current hard disk operating mode and the current target power consumption value determined in step S206, and the corresponding relationship between the hard disk power supply parameter and the power consumption value and operating mode determined in step S204. And finally, writing the current target power supply parameter into a main control chip of the SSD through a Set Feature (power supply management interface) provided by an NVME protocol, so as to realize the final dynamic management of the SSD power supply state and achieve the purpose of managing the power consumption.
And S208, writing the current target power supply parameter into a main control chip of the hard disk.
Similar to the aforementioned step S104, the description is omitted here.
S209, obtaining the current actual power supply parameter through a power supply management interface provided by the NVME protocol so as to realize the closed loop of the hard disk power supply management.
In the embodiment of the invention, the BMC can acquire the current data power supply parameter of the hard disk according to the Get Feature command in the power supply management interface provided by the NVME protocol so as to realize the closed loop of dynamic management and realize accurate control. Specifically, the BMC can determine the correctness of the Set Feature through the Get Feature, and further accurately regulate and control the parameters in real time.
By adopting the hard disk power management method provided by the embodiment of the invention, the BMC can complete the model training step and the data analysis step in advance to respectively obtain the target power consumption prediction model and the corresponding relation between the hard disk power parameters, the power consumption values and the working modes. Further, the optimal hard disk power supply parameters can be matched according to the client running scene automatically identified in real time, so that the hard disk power supply parameters are dynamically controlled, and the dynamic management of the hard disk power supply state is completed. Therefore, the hard disk does not need to operate in a high-power-consumption working mode all the time, power consumption is wasted, over-high power consumption is poor in scattering, the service life of the SSD is damaged, and performance requirements of a client operation scene cannot be met and client experience is lost because the hard disk operates in a low-power-consumption working mode.
Referring to fig. 3, an interaction flow diagram of a hard disk power management method according to an embodiment of the present invention is shown. Specifically, in the embodiment of the present invention, the BMC may acquire a plurality of client operation scenes in the server operation process, and operate the decision tree algorithm, and in practical application, the BMC automatically identifies the plurality of client scenes, and acquires the data read-write delay parameter and the power consumption parameter in various scenes, so as to determine the best matching between the service performance requirement of the operable scenes and the power state, that is, the matching between the hard disk data read-write delay parameter and the power consumption parameter, in different client scenes according to the operation result of the decision tree algorithm. Further, the BMC bottom layer determines corresponding hard disk Power supply parameters (parameters such as Entry Latency, exit Latency, and the like in the Power State Descriptor structure of the hard disk provided by the NVEM protocol) according to the Power consumption value and the working State of the hard disk, and writes the hard disk Power supply parameters into the SSD main control chip, so that the SSD main control chip controls the working State of the hard disk Power supply, and further, the SSD main control may also obtain actual Power supply parameters of the hard disk, and feed back the actual Power supply parameters to the BMC bottom layer through Get Feature commands in the Power management interface, so as to complete the closed loop of hard disk Power supply management, and implement accurate control.
Therefore, in the embodiment of the invention, a method for dynamically managing the power state of the NVME hard disk by the BMC based on machine learning of the decision tree algorithm is provided, which can automatically identify the operation scene of a client, ensure that the service requirement is met, and simultaneously give the optimal power consumption strategy, thereby realizing dynamic management of the working state of the SSD hard disk of the NVME protocol. In the embodiment of the invention, a whole set of decision tree of the power consumption strategy of the NVME protocol hard disk is constructed, and the switching of the SSD power consumption strategy can be spontaneously completed by monitoring the BMC serving as the Host terminal.
Referring to fig. 4, a flowchart of a hard disk power management method according to an embodiment of the present invention is shown. Specifically, the hard disk power management method provided by the embodiment of the present invention may include the following steps:
s401, receiving the current target power supply parameter written by the BMC through a power supply management interface provided by the NVME protocol.
In the embodiment of the invention, the main control chip of the hard disk can provide a power management interface based on the NVME protocol, so that the target power parameter written by the BMC can be received based on the power management interface, and the dynamic management of the power state of the hard disk controlled by the BMC is realized.
S402, managing the power state of the hard disk according to the current target power parameter.
In the embodiment of the invention, after the main control chip of the hard disk receives the target power supply parameter written by the BMC, the Set Feature command of the BMC can be executed so as to control the power supply state of the hard disk.
The BMC is used for acquiring current data read-write delay parameters and determining a current client operation scene; determining a current hard disk working mode and a current target power consumption value according to the current client running scene and a pre-trained target power consumption prediction model, wherein the target power consumption prediction model learns the corresponding relation among the client running scene, the hard disk working mode and the target power consumption value; and determining the current target power supply parameter according to the current hard disk working mode and the current target power consumption value.
In the embodiment of the invention, the main control chip of the hard disk can also send historical hard disk power supply parameters, corresponding historical power consumption values and historical working modes to the BMC through a power supply management interface provided by the NVEM protocol.
The BMC is further used for acquiring historical hard disk power supply parameters, corresponding historical power consumption values and historical working modes through a power supply management interface provided by the NVEM protocol, taking the historical hard disk power supply parameters as independent variables, taking the corresponding historical power consumption values and the corresponding historical working modes as dependent variables, and determining the corresponding relation among the hard disk power supply parameters, the power consumption values and the working modes.
In the embodiment of the invention, the main control chip of the hard disk can send the real-time hard disk power supply parameters, the corresponding power consumption value and the working mode to the BMC through the power supply management interface provided by the NVEM protocol so that the BMC can complete data analysis and determine the corresponding relation between the hard disk power supply parameters, the power consumption value and the working mode.
Referring to fig. 5, a flowchart of a hard disk power management method according to an embodiment of the present invention is shown. Specifically, the hard disk power management method provided by the embodiment of the present invention may include the following steps:
s501, receiving the current target power supply parameter written by the BMC through a power supply management interface provided by the NVME protocol.
And S502, managing the power state of the hard disk according to the current target power parameter.
Steps S501 to S502 are similar to steps S401 to S402, and the embodiment of the present invention is not described herein again.
And S503, acquiring the current actual power supply parameter.
S504, the current actual power supply parameter is fed back to the BMC, so that the BMC can obtain the current actual power supply parameter, and closed loop of hard disk power supply management is achieved.
In the embodiment of the invention, the SSD master control chip can acquire the current actual power supply parameters and feed back the current actual power supply parameters to the BMC so as to realize the closed loop of hard disk power supply management.
Hereinafter, the hard disk power management method provided by the embodiment of the present invention will be described with reference to a specific example of an interaction process between the BMC and the hard disk power management device, where the example is only used to explain the technical solution provided by the present invention, and does not limit the hard disk power management method provided by the embodiment of the present invention.
Illustratively, the BMC needs to complete two preparation tasks, including:
1. the method comprises the steps of obtaining various parameters of a client operation scene (including videos, various databases, financial transactions, the Internet of things and the like), taking data of each client operation scene as a sample, and training a preset model based on a decision tree algorithm to obtain a target power consumption prediction model.
2. The power management interface provided by the NVME protocol is used for acquiring the power parameter data of the hard disk, and the corresponding power consumption data and working mode data, so that the analysis of the related data of the hard disk power supply can be completed, and the corresponding relation among the power parameter, the power consumption value and the working mode of the hard disk can be determined.
After the two preparation works are completed, the BMC can dynamically manage the hard disk power supply based on the results of the two preparation works, the current operation scene is assumed to be a video scene, the data read-write delay parameter is high, the hard disk is required to operate in a high-power-consumption working mode, after the BMC determines the data read-write delay condition, the BMC can determine a corresponding target power consumption value and a working mode based on a pre-trained target power consumption prediction model, further determine a target hard disk power supply parameter based on the corresponding relation between the predetermined hard disk power supply parameter, the power consumption value and the working mode, write the target hard disk power supply parameter into the SSD main control chip through a Set Feature command, after the SSD main control chip receives the command, execute the command, control the hard disk power supply state, and feed back the actual hard disk power supply state to the BMC through the Set Feature so as to realize the closed loop of the dynamic management and realize accurate control.
Therefore, in the embodiment of the present invention, the BMC may write the current target power parameter into the main control chip of the hard disk through the power management interface provided by the NVEM protocol, so that the main control chip controls the SSD power state, and dynamic management of the SSD power state is implemented.
Based on the same inventive concept, an embodiment of the present invention provides a hard disk power management device, which is applied to a BMC, and fig. 6 is a schematic diagram of the hard disk power management device provided in the embodiment of the present invention, as shown in fig. 6, the hard disk power management device includes:
a first determining module 601, configured to collect current data read-write delay parameters and determine a current client operation scene;
a second determining module 602, configured to determine a current hard disk operating mode and a current target power consumption value according to the current client operating scenario and a pre-trained target power consumption prediction model, where the target power consumption prediction model learns a corresponding relationship between the client operating scenario, the hard disk operating mode, and the target power consumption value;
a third determining module 603, configured to determine a current target power supply parameter according to the current hard disk operating mode and the current target power consumption value;
and a write-in module 604, configured to write the current target power parameter into a main control chip of a hard disk.
Optionally, the apparatus further comprises:
the system comprises a marking module, a processing module and a processing module, wherein the marking module is used for obtaining a sample client operation scene, marking the sample client operation scene and obtaining a target power consumption value label corresponding to the sample client operation scene;
the training module is used for performing model training on the read-write delay characteristic, the hard disk working mode characteristic and the target power consumption value label of the sample client operation scene by adopting a decision tree algorithm to obtain a target power consumption prediction model;
the working mode is characterized by the proportion of two access modes, namely sequential read-write access and random read-write access.
Optionally, the apparatus further comprises:
the acquisition module is used for acquiring historical hard disk power supply parameters, corresponding historical power consumption values and historical working modes through a power supply management interface provided by an NVEM protocol;
the fourth determining module is used for determining the corresponding relation among the power supply parameters, the power consumption values and the working modes of the hard disk by taking the historical power supply parameters as independent variables and the corresponding historical power consumption values and the historical working modes as dependent variables;
the third determining module 603 is specifically configured to:
and determining the current target power supply parameter according to the current hard disk working mode, the current target power consumption value and the corresponding relation between the hard disk power supply parameter, the power consumption value and the working mode.
Optionally, the apparatus further comprises:
and the acquisition module is used for acquiring the current actual power supply parameters through a power supply management interface provided by the NVME protocol so as to realize the closed loop of the hard disk power supply management.
Based on the same inventive concept, an embodiment of the present invention provides a hard disk power management apparatus, which is applied to a main control chip of a hard disk, and fig. 7 is a schematic diagram of the hard disk power management apparatus provided in the embodiment of the present invention, as shown in fig. 7, the apparatus includes:
a receiving module 701, configured to receive, through a power management interface provided by the NVME protocol, a current target power parameter written by the BMC;
a management module 702, configured to manage a power state of the hard disk according to the current target power parameter;
the BMC is used for acquiring current data read-write delay parameters and determining a current client operation scene; determining a current hard disk working mode and a current target power consumption value according to the current client running scene and a pre-trained target power consumption prediction model, wherein the target power consumption prediction model learns the corresponding relation among the client running scene, the hard disk working mode and the target power consumption value; and determining the current target power supply parameter according to the current hard disk working mode and the current target power consumption value.
Optionally, the apparatus further comprises:
the acquisition module is used for acquiring the current actual power supply parameters;
and the feedback module is used for feeding the current actual power supply parameter back to the BMC so that the BMC can acquire the current actual power supply parameter and realize the closed loop of hard disk power supply management.
Optionally, the apparatus further comprises:
the sending module is used for sending the historical hard disk power supply parameters, the corresponding historical power consumption values and the historical working modes to the BMC through a power supply management interface provided by an NVEM protocol;
the BMC is further used for acquiring historical hard disk power supply parameters, corresponding historical power consumption values and historical working modes through a power supply management interface provided by the NVEM protocol, taking the historical hard disk power supply parameters as independent variables, taking the corresponding historical power consumption values and the corresponding historical working modes as dependent variables, and determining the corresponding relation among the hard disk power supply parameters, the power consumption values and the working modes.
Based on the same inventive concept, an embodiment of the present invention provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the steps of the hard disk power management method according to any one of the above embodiments of the present invention.
Based on the same inventive concept, embodiments of the present invention provide a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the steps of the hard disk power management method according to any of the above embodiments of the present invention.
Based on the same inventive concept, embodiments of the present invention provide a computer program product, which includes a computer program/instruction, and when the computer program/instruction is executed by a processor, the steps of the hard disk power management method according to any of the above embodiments of the present invention are implemented.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications of these embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the invention.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "include", "including" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or terminal device including a series of elements includes not only those elements but also other elements not explicitly listed or inherent to such process, method, article, or terminal device. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or terminal apparatus that comprises the element.
The hard disk power management method, device, electronic device, medium and product provided by the invention are introduced in detail, and a specific example is applied in the text to explain the principle and the implementation of the invention, and the description of the above embodiment is only used to help understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (12)
1. A hard disk power management method is applied to BMC and comprises the following steps:
acquiring current data read-write delay parameters and determining a current client operation scene;
determining a current hard disk working mode and a current target power consumption value according to the current client running scene and a pre-trained target power consumption prediction model, wherein the target power consumption prediction model learns the corresponding relation among the client running scene, the hard disk working mode and the target power consumption value;
determining a current target power supply parameter according to the current hard disk working mode and the current target power consumption value;
and writing the current target power supply parameter into a main control chip of the hard disk.
2. The method of claim 1, further comprising:
obtaining a sample client operation scene, labeling the sample client operation scene, and obtaining a target power consumption value label corresponding to the sample client operation scene;
performing model training on the read-write delay characteristic, the hard disk working mode characteristic and the target power consumption value label of the sample client operation scene by adopting a decision tree algorithm to obtain a target power consumption prediction model;
the working mode is characterized by the proportion of two access modes, namely sequential read-write access and random read-write access.
3. The method of claim 1, further comprising:
acquiring historical hard disk power supply parameters, corresponding historical power consumption values and historical working modes through a power supply management interface provided by an NVEM (network video and Environment) protocol;
taking the historical hard disk power supply parameters as independent variables, and taking corresponding historical power consumption values and historical working modes as dependent variables, and determining the corresponding relation among the hard disk power supply parameters, the power consumption values and the working modes;
determining a current target power supply parameter according to the current hard disk working mode and the current target power consumption value, wherein the determining comprises the following steps:
and determining the current target power supply parameter according to the current hard disk working mode, the current target power consumption value and the corresponding relation between the hard disk power supply parameter, the power consumption value and the working mode.
4. The method of claim 1, wherein after writing the current target power supply parameter to a master chip of the SSD, the method further comprises:
and acquiring the current actual power supply parameters through a power supply management interface provided by the NVME protocol so as to realize the closed loop of the hard disk power supply management.
5. A hard disk power supply management method is characterized in that the method is applied to a main control chip of a hard disk, and the method comprises the following steps:
receiving a current target power supply parameter written by the BMC through a power supply management interface provided by an NVME protocol;
managing the power state of the hard disk according to the current target power parameter;
the BMC is used for acquiring current data read-write delay parameters and determining a current client operation scene; determining a current hard disk working mode and a current target power consumption value according to the current client running scene and a pre-trained target power consumption prediction model, wherein the target power consumption prediction model learns the corresponding relation among the client running scene, the hard disk working mode and the target power consumption value; and determining the current target power supply parameter according to the current hard disk working mode and the current target power consumption value.
6. The method of claim 5, wherein after managing the power state of the hard disk according to the current target power parameter, the method further comprises:
acquiring current actual power supply parameters;
and feeding back the current actual power supply parameter to the BMC so that the BMC can acquire the current actual power supply parameter and realize the closed loop of hard disk power supply management.
7. The method of claim 5, further comprising:
sending historical hard disk power supply parameters, corresponding historical power consumption values and historical working modes to the BMC through a power management interface provided by an NVEM (network video and Environment) protocol;
the BMC is further used for acquiring historical hard disk power supply parameters, corresponding historical power consumption values and historical working modes through a power supply management interface provided by the NVEM protocol, taking the historical hard disk power supply parameters as independent variables, taking the corresponding historical power consumption values and the corresponding historical working modes as dependent variables, and determining the corresponding relation among the hard disk power supply parameters, the power consumption values and the working modes.
8. A hard disk power management device is applied to BMC, and the device comprises:
the first determining module is used for acquiring current data read-write delay parameters and determining a current client operation scene;
the second determination module is used for determining a current hard disk working mode and a current target power consumption value according to the current client running scene and a pre-trained target power consumption prediction model, and the target power consumption prediction model learns the corresponding relation among the client running scene, the hard disk working mode and the target power consumption value;
the third determining module is used for determining the current target power supply parameter according to the current hard disk working mode and the current target power consumption value;
and the writing module is used for writing the current target power supply parameter into a main control chip of a hard disk.
9. The utility model provides a hard disk power management device which characterized in that is applied to the main control chip of hard disk, the device includes:
the receiving module is used for receiving the current target power supply parameter written by the BMC through a power supply management interface provided by the NVME protocol;
the management module is used for managing the power state of the hard disk according to the current target power parameter;
the BMC is used for acquiring current data read-write delay parameters and determining a current client operation scene; determining a current hard disk working mode and a current target power consumption value according to the current client running scene and a pre-trained target power consumption prediction model, wherein the target power consumption prediction model learns the corresponding relation among the client running scene, the hard disk working mode and the target power consumption value; and determining the current target power supply parameter according to the current hard disk working mode and the current target power consumption value.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the steps of the hard disk power management method of any one of claims 1-4 or 5-7 are implemented when the program is executed by the processor.
11. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the hard disk power management method according to any one of claims 1 to 4 or 5 to 7.
12. A computer program product comprising computer programs/instructions, characterized in that the computer programs/instructions, when executed by a processor, implement the steps of the hard disk power management method according to any of claims 1-4 or 5-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211261792.9A CN115826729A (en) | 2022-10-14 | 2022-10-14 | Hard disk power management method, device, electronic equipment, medium and product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211261792.9A CN115826729A (en) | 2022-10-14 | 2022-10-14 | Hard disk power management method, device, electronic equipment, medium and product |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115826729A true CN115826729A (en) | 2023-03-21 |
Family
ID=85524788
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211261792.9A Pending CN115826729A (en) | 2022-10-14 | 2022-10-14 | Hard disk power management method, device, electronic equipment, medium and product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115826729A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116520969A (en) * | 2023-07-04 | 2023-08-01 | 四川弘仁财电科技有限公司 | Device, system and method for controlling power consumption of server |
-
2022
- 2022-10-14 CN CN202211261792.9A patent/CN115826729A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116520969A (en) * | 2023-07-04 | 2023-08-01 | 四川弘仁财电科技有限公司 | Device, system and method for controlling power consumption of server |
| CN116520969B (en) * | 2023-07-04 | 2023-09-08 | 四川弘仁财电科技有限公司 | Device, system and method for controlling power consumption of server |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108416003B (en) | Picture classification method and device, terminal and storage medium | |
| DE112011100143B4 (en) | Optimize electrical power consumption in a data center | |
| US20230066050A1 (en) | Image classification modeling while maintaining data privacy compliance | |
| US11784939B2 (en) | Cloud resource management using externally-sourced data | |
| US10902339B2 (en) | System and method providing automatic completion of task structures in a project plan | |
| CN106708443B (en) | Data reading and writing method and device | |
| US20210216904A1 (en) | Knowledge Aided Feature Engineering | |
| US20200074509A1 (en) | Business data promotion method, device, terminal and computer-readable storage medium | |
| CN112732949B (en) | Service data labeling method and device, computer equipment and storage medium | |
| US11594054B2 (en) | Document lineage management system | |
| CN112163614A (en) | Anchor classification method and device, electronic equipment and storage medium | |
| CN115826729A (en) | Hard disk power management method, device, electronic equipment, medium and product | |
| CN111291564B (en) | Model training method, device and storage medium for word vector acquisition | |
| CN115729965A (en) | Information stream processing method, device, stream server and storage medium | |
| CN116126415A (en) | Scene-based cloud component configuration recommendation method and system | |
| CN111538859B (en) | Method and device for dynamically updating video tag and electronic equipment | |
| CN114676272A (en) | Information processing method, device and equipment of multimedia resource and storage medium | |
| CN109062396B (en) | Method and device for controlling equipment | |
| CN113703986B (en) | Information management system and method based on big data | |
| CN111091198A (en) | Data processing method and device | |
| US20230169408A1 (en) | Annotation of a Machine Learning Pipeline with Operational Semantics | |
| CN115048999A (en) | Label optimization method and device, electronic equipment and readable medium | |
| KR20230087308A (en) | Cloud learning device and method through artificial intelligence-based resource utilization pattern prediction | |
| Wang et al. | The pure exploration problem with general reward functions depending on full distributions | |
| CN117726917A (en) | Model training method, device, electronic equipment and computer storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |