CN115373858B - Peak-shifting adjusting method and system based on constant power of calculating part - Google Patents
Peak-shifting adjusting method and system based on constant power of calculating part Download PDFInfo
- Publication number
- CN115373858B CN115373858B CN202211315622.4A CN202211315622A CN115373858B CN 115373858 B CN115373858 B CN 115373858B CN 202211315622 A CN202211315622 A CN 202211315622A CN 115373858 B CN115373858 B CN 115373858B
- Authority
- CN
- China
- Prior art keywords
- core
- power
- frequency
- idle
- current
- 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.)
- Active
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
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/505—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the load
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/508—Monitor
-
- 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
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Power Sources (AREA)
Abstract
The invention provides a peak shifting adjusting method and system based on constant power of a calculating part, belonging to the technical field of resource allocation, wherein the peak shifting adjusting method comprises the following steps: obtaining the rated power and the number of cores of the calculation part, and distributing a power threshold value to each core: uploading the activity frequency of each kernel of the computing part to a time slice; calculating the current operating power of each core according to the active frequency of each core of the time slice; and analyzing the load operation condition and the resource request quantity of the current computing part according to the current operation power and the power threshold value of each core, and adjusting the frequency multiplication coefficient of each core by using the BIOS. The invention adjusts the frequency multiplication coefficient of each core through the BIOS so as to realize the power allocation among the cores, and can ensure that the computing part automatically carries out frequency modulation, thereby not only improving the running speed of the computing part, but also ensuring the effective utilization of resources, prolonging the service life of the computing part and improving the stability of the whole system.
Description
Technical Field
The invention relates to the technical field of resource allocation, in particular to a peak staggering regulation method and system based on constant power of a calculation part.
Background
With the development of the application of the integrated and intelligent technology, the computing part is used as the core of operation and control, and the operation speed and the resource allocation processing capacity play a decisive role in the whole system. The resource request amount of each execution unit to the computing unit is influenced by a series of factors such as holidays, morning and evening peaks and the like, and real-time performance and uncertainty exist. For example: in the morning and evening peak of the subway, along with the increase of the passenger flow, the resource requests of part of execution units to the calculation part are multiplied, and when the large and complex requests are responded, if the large and complex requests cannot be processed in time, the execution efficiency of the execution units is influenced, so that a series of continuous problems are caused.
At present, in order to cope with the situation of sudden increase of the burst request amount, the highest frequency which can be reached by the calculation part is usually locked, so that the calculation part continuously operates at the highest frequency. Although the performance of the computing part can be improved by the method, the technical scheme can cause the computing part to operate in a state of far exceeding the rated frequency for a long time, so that the power consumption of the computing part is increased and the service life of the computing part is reduced.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention aims to provide a peak-shifting adjusting method and system based on constant power of a calculating part.
The invention provides a peak shift adjusting method based on constant power of a calculating part, which comprises the following steps:
step 1: acquiring rated power of a calculation part and the number of calculation units, and distributing a power threshold value to each calculation unit;
step 2: uploading the activity frequency of each computing unit to a time slice;
and step 3: calculating the current operating power of each core according to the activity frequency of each calculating unit of the time slice;
and 4, step 4: and analyzing the load operation condition and the resource request quantity of the current computing part according to the current operation power and the power threshold value of each computing unit, and adjusting the frequency multiplication coefficient of each computing unit by using the BIOS.
Preferably, the calculation unit may be each core of one calculation unit, or each calculation unit, GPU or TPU in a computer composed of multiple calculation units.
Preferably, the step 3: calculating the current operating power of each core according to the active frequency of each core of the time slice, wherein the method comprises the following steps:
the formula is adopted:
calculating the current operating power of each core; wherein, G is a power threshold value, ps is the current core operating power, fs is an active frequency, fn is the rated frequency of each core of the calculation part, us is the current voltage of the calculation part, and Un is the rated voltage of the calculation part.
Preferably, the step 4: analyzing the load operation condition and the resource request quantity of the current computing part according to the current operation power and the power threshold value of each kernel, and adjusting the frequency multiplication coefficient of each kernel by using the BIOS, wherein the method comprises the following steps:
step 4.1: performing FFT (fast Fourier transform) on the activity frequency of each core in two adjacent time slices to obtain the rising and falling trend of the activity frequency of each core;
step 4.2: extracting an idle kernel with a descending activity frequency, and calculating the idle power of the idle kernel;
step 4.3: and adjusting the frequency multiplication coefficient of each core by using the BIOS, and transferring the idle power of the idle core to the target core.
Preferably, the step 4: analyzing the load operation condition and the resource request quantity of the current computing part according to the current operation power and the power threshold value of each core, and adjusting the frequency multiplication coefficient of each core by using a BIOS (basic input output System), wherein the method comprises the following steps:
and performing FFT (fast Fourier transform) on the active frequency of each core, judging whether the active frequency of the cores in the plurality of time slices is linearly increased or not, closing the corresponding idle core when the active frequency of the cores is not linearly increased, adjusting the frequency multiplication coefficient of each core by using the BIOS, and transferring the idle power of the idle core to the target core.
The invention also provides a peak shift adjusting system based on the constant power of the calculating part, which comprises the following components:
the power distribution module is used for acquiring the rated power and the number of the cores of the calculation part and distributing a power threshold value for each core;
the activity frequency uploading module is used for uploading the activity frequency of each kernel of the computing part to the time slice;
the running power calculation module is used for calculating the current running power of each core according to the activity frequency of each core of the time slice;
and the kernel adjusting module is used for analyzing the load operation condition and the resource request quantity of the current computing part according to the current operation power and the power threshold value of each kernel and adjusting the frequency multiplication coefficient of each kernel by using the BIOS.
Preferably, the computing unit includes a computing unit or a computer, and the computing unit may be each core of one computing unit, or each computing unit, GPU or TPU in a computer composed of multiple computing units.
Preferably, the operating power calculating module includes:
the formula is adopted:
calculating the current operating power of each core; wherein, G is a power threshold value, ps is the current core operating power, fs is an active frequency, fn is the rated frequency of each core of the calculation part, us is the current voltage of the calculation part, and Un is the rated voltage of the calculation part.
Preferably, the kernel adjusting module includes:
the FFT conversion unit is used for carrying out FFT conversion on the activity frequency of each core in two adjacent time slices to obtain the rising and falling trend of the activity frequency of each core;
the idle power calculation module is used for extracting an idle kernel with a descending activity frequency and calculating the idle power of the idle kernel;
and the power transfer module is used for adjusting the frequency multiplication coefficient of each core by using the BIOS and transferring the idle power of the idle core to the target core.
Preferably, the kernel adjusting module further includes:
and the idle kernel adjusting unit is used for carrying out FFT (fast Fourier transform) on the active frequency of each kernel, judging whether the active frequency of the kernels in the time slices is linearly increased or not, closing the corresponding idle kernel when the active frequency of the kernels is not linearly increased, adjusting the frequency multiplication coefficient of each kernel by using the BIOS (basic input output system), and transferring the idle power of the idle kernel to a target kernel.
The method and the system for adjusting the peak shift based on the constant power of the calculation part have the advantages that: compared with the prior art, the invention analyzes the load operation condition and the resource request quantity of the current computing part by utilizing the current operation power and the power threshold value of each inner core, and adjusts the frequency multiplication coefficient of each inner core through the BIOS so as to realize the power allocation among the cores, so that the computing part can automatically carry out frequency modulation, the operation speed of the computing part is improved, the effective utilization of resources is ensured, the service life of the computing part is prolonged, and the stability of the whole system is improved.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of a peak shift adjustment method based on constant power of a calculation part according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating the peak shift adjustment of the constant power of the calculation unit according to the embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a frequency modulation scheme of a computing portion according to an embodiment of the present invention;
fig. 4 shows a schematic diagram of core power adjustment provided by an embodiment of the present invention.
Detailed Description
In the description of the present invention, it is to be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", and the like, indicate orientations and positional relationships based on those shown in the drawings, and are used only for convenience of description and simplicity of description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be considered as limiting the present invention.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present invention, "a plurality" means two or more unless specifically defined otherwise.
In the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," "secured," and the like are to be construed broadly and can, for example, be fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
Referring to fig. 1-2, the present invention provides a peak shift adjusting method based on constant power of a calculating part, including:
step 1: acquiring the rated power and the number of cores of the calculation part, and distributing a power threshold value for each core;
in practical application, a server is started, BIOS analyzes basic parameter configuration of the server after starting operation, and parameters such as rated power (Pn) of a computing part and activity frequency of each kernel are converted into corresponding binary digital signals. Reading the number (N) of cores of the calculation part and distributing a power threshold value (G) for each core, wherein the calculation formula is as follows; pn = N × G, pn is the calculation unit rated power, N is the number of cores, and G is the core power threshold.
Step 2: uploading the activity frequency of each kernel of the computing part to a time slice;
it should be noted that the activity frequency of each core of the computing unit of the present invention is recorded in a time slice manner, the computing unit includes a computing unit or a computer, and the computing unit may be each core of one computing unit, or each computing unit, GPU or TPU in a computer composed of multiple computing units. In practical application, the latest core activity frequency is uploaded to a time sliceThen, the recorded real-time activity frequency in the time slice is subjected to maximum and minimum values removal, and then FFT (Fourier transform) calculation is carried out. Thereby obtaining the reliable activity frequency of each kernel of the calculation part in the set time slice. The FFT calculation method comprises the following steps: time shifting: f [ X (t-to)] = X(jw)*(ii) a Frequency shift: f (X (t)) = X[j(w + wo)]. Wherein t is the time slice ending time, to is the time slice starting time, w is the angular frequency, and wo is the initial angular frequency.
And step 3: calculating the current operating power of each core according to the active frequency of each core of the time slice;
further, the step 3 comprises:
the formula is adopted:
calculating the current operating power of each core; wherein, G is a power threshold value, ps is the current core operating power, fs is an active frequency, fn is the rated frequency of each core of the calculation part, us is the current voltage of the calculation part, and Un is the rated voltage of the calculation part.
And 4, step 4: and analyzing the load operation condition and the resource request quantity of the current computing part according to the current operation power and the power threshold value of each core, and adjusting the frequency multiplication coefficient of each core by using the BIOS.
The invention provides a method for comparing the running power (Ps) of each core of a computing part with a power threshold (G), which can analyze the load running condition and the resource request quantity of the current computing part, then adjust the frequency multiplication coefficient of each core by using a BIOS (basic input/output System), further adjust the running power of each core, complete the resource initialization allocation work and enable the computing part to run at the most appropriate frequency.
Further, the step 4 comprises:
step 4.1: performing FFT (fast Fourier transform) on the activity frequency of each core in two adjacent time slices to obtain the rising and falling trend of the activity frequency of each core;
step 4.2: extracting an idle kernel with a descending activity frequency, and calculating the idle power of the idle kernel;
step 4.3: and adjusting the frequency multiplication coefficient of each core by using the BIOS, and transferring the idle power of the idle core to the target core.
Referring to fig. 3, in practical applications, when a core of a computing unit encounters a sudden increase in the amount of burst requests, the core needs the computing unit to provide a stronger performance for fast computing processing:
(1) According to the activity frequency of each core of the time slice, the current running power (Ps) of each core is calculated, and the calculation formula is as follows:
g is a power threshold value, ps is the current core operating power, fs is the real-time activity frequency, fn is the rated frequency of each core of the calculation part, us is the current voltage of the calculation part, and Un is the rated voltage of the calculation part;
(2) Time-shift characteristic F [ X (t-to) by FFT transformation] = X(jw)*Performing FFT (fast Fourier transform) on the activity frequency of each core in two adjacent time slices, analyzing the ascending and descending trend of the activity frequency of each core according to the activity frequency calculated in the two adjacent time slices, calculating the idle power of the core with the activity frequency becoming the descending trend: pi = G (power threshold) -Ps (current kernel operating power), the frequency multiplication coefficient of each kernel is adjusted through the BIOS, the idle power of the kernel is transferred to the kernel with larger operating power, the power of the operating kernel is improved, the over-frequency operation is realized, and the current task is completed more quickly. Note: at operating power ratio = core operating power/rated power.
Further, step 4 further comprises: and performing FFT (fast Fourier transform) on the active frequency of each core, judging whether the active frequency of the cores in the plurality of time slices is linearly increased or not, closing the corresponding idle core when the active frequency of the cores is not linearly increased, adjusting the frequency multiplication coefficient of each core by using the BIOS, and transferring the idle power of the idle core to the target core.
Referring to fig. 4, in practical applications, when the calculating portion is in a relatively idle state: according to the invention, the core active frequency with low operating power and active frequency is subjected to FFT conversion through the time shift characteristic of the FFT conversion, and the core active frequency is extremely low and the active frequency is subjected to wireless increase fluctuation in a plurality of time slices, so that the core is considered to be in an idle state temporarily. The calculation part sets the frequency multiplication coefficient of the BIOS through analysis and judgment, intelligently closes the corresponding idle core, and moves the power for closing the core to the running core, so that the over-frequency of the running core is realized, the computer can more quickly complete the current task, and the calculation resources are saved.
The invention allocates the rated power (Pn) of the calculating part to each core without increasing the power of the calculating part, and takes the power value allocated to each core as the core power threshold value (G). The calculation unit analyzes the operating power (Ps) of each core and compares the operating power (Ps) with a power threshold (G) allocated to each core. If one or more cores have current operating power which does not reach the power threshold (G) of the corresponding core, the invention can adjust the frequency multiplication coefficient of each core through the BIOS and move the idle power (Pu) to the core which operates at or above the power threshold, thereby realizing automatic frequency modulation, improving the processing speed and completing the adjustment of the operation resources.
The invention also provides a peak-shifting regulation system based on the constant power of the calculation part, which comprises the following components:
the power distribution module is used for acquiring the rated power and the number of the cores of the calculation part and distributing a power threshold value for each core;
the activity frequency uploading module is used for uploading the activity frequency of each kernel of the computing part to the time slice;
the running power calculation module is used for calculating the current running power of each core according to the activity frequency of each core of the time slice;
and the kernel adjusting module is used for analyzing the load operation condition and the resource request quantity of the current computing part according to the current operation power and the power threshold value of each kernel and adjusting the frequency multiplication coefficient of each kernel by using the BIOS.
Preferably, the operating power calculating module includes:
the formula is adopted:
calculating the current operating power of each core; wherein, G is a power threshold value, ps is the current core operating power, fs is an active frequency, fn is the rated frequency of each core of the calculation part, us is the current voltage of the calculation part, and Un is the rated voltage of the calculation part.
Preferably, the kernel adjusting module includes:
the FFT conversion unit is used for carrying out FFT conversion on the activity frequency of each core in two adjacent time slices to obtain the rising and falling trend of the activity frequency of each core;
the idle power calculation module is used for extracting an idle kernel with the activity frequency of falling trend and calculating the idle power of the idle kernel;
and the power transfer module is used for adjusting the frequency multiplication coefficient of each core by using the BIOS and transferring the idle power of the idle core to the target core.
Preferably, the kernel adjusting module further includes:
and the idle core adjusting unit is used for carrying out FFT (fast Fourier transform) on the active frequency of each core, judging whether the active frequency of the cores in the plurality of time slices is linearly increased or not, closing the corresponding idle core when the active frequency of the cores is not linearly increased, adjusting the frequency multiplication coefficient of each core by using the BIOS (basic input/output system), and transferring the idle power of the idle core to the target core.
The method and the system for adjusting the peak shift based on the constant power of the calculation part have the advantages that: compared with the prior art, the invention adjusts the frequency through the BIOS under the condition of ensuring that the overall power of the computing part is constant, adjusts the operating power of each core of the computing part, distributes the cores of the computing part to the most appropriate active frequency in real time, and realizes the automatic frequency modulation of the computing part, thereby improving the operating speed of the computing part, ensuring the effective utilization of resources, prolonging the service life of the computing part and improving the stability of the whole system.
The present invention also provides a computer-readable storage medium, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the steps in a method for peak shift adjustment based on constant power of a calculation section as described above.
Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the embodiment of the invention are the same as the beneficial effects of the peak shifting adjusting method based on the constant power of the calculating part in the technical scheme, and are not described herein again.
The above description is only an embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of changes or substitutions within the technical scope of the present invention, and the present invention shall be covered by the claims. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (4)
1. A peak shift adjustment method based on constant power of a calculation part, wherein the calculation part comprises a plurality of calculation units, and is characterized by comprising the following steps:
step 1: acquiring rated power of a calculating part and the number of calculating units, and distributing a power threshold value to each calculating unit;
step 2: uploading the activity frequency of each computing unit to a time slice;
and step 3: calculating the current operating power of each core according to the activity frequency of each calculating unit of the time slice;
and 4, step 4: analyzing the load operation condition and the resource request quantity of the current computing part according to the current operation power and the power threshold value of each computing unit, and adjusting the frequency multiplication coefficient of each computing unit by using a BIOS (basic input output System);
the computing part comprises a computing part or a computer, and the computing unit can be each core of one computing part or each computing part, GPU or TPU in the computer composed of a plurality of computing parts;
the step 3: calculating the current operating power of each core according to the active frequency of each core of the time slice, wherein the method comprises the following steps:
the formula is adopted:
calculating the current operating power of each core; g is a power threshold value, ps is the current core operating power, fs is an active frequency, fn is the rated frequency of each core of the calculation part, us is the current voltage of the calculation part, and Un is the rated voltage of the calculation part;
the step 4: analyzing the load operation condition and the resource request quantity of the current computing part according to the current operation power and the power threshold value of each kernel, and adjusting the frequency multiplication coefficient of each kernel by using the BIOS, wherein the method comprises the following steps:
step 4.1: performing FFT (fast Fourier transform) on the activity frequency of each core in two adjacent time slices to obtain the rising and falling trend of the activity frequency of each core;
step 4.2: extracting an idle kernel with a descending activity frequency, and calculating the idle power of the idle kernel;
step 4.3: and adjusting the frequency multiplication coefficient of each core by using the BIOS, and transferring the idle power of the idle core to the target core.
2. The peak shift adjusting method based on constant power of the calculating part according to claim 1, wherein the step 4: analyzing the load operation condition and the resource request quantity of the current computing part according to the current operation power and the power threshold value of each kernel, and adjusting the frequency multiplication coefficient of each kernel by using the BIOS, wherein the method comprises the following steps:
and performing FFT (fast Fourier transform) on the active frequency of each core, judging whether the active frequency of the cores in the plurality of time slices is linearly increased or not, closing the corresponding idle core when the active frequency of the cores is not linearly increased, adjusting the frequency multiplication coefficient of each core by using the BIOS, and transferring the idle power of the idle core to the target core.
3. A peak shift adjustment system based on constant power of a calculation part is characterized by comprising:
the power distribution module is used for acquiring the rated power and the number of the cores of the calculation part and distributing a power threshold value for each core;
the activity frequency uploading module is used for uploading the activity frequency of each kernel of the computing part to the time slice;
the running power calculation module is used for calculating the current running power of each core according to the activity frequency of each core of the time slice;
the kernel adjusting module is used for analyzing the load operation condition and the resource request quantity of the current computing part according to the current operation power and the power threshold value of each kernel and adjusting the frequency multiplication coefficient of each kernel by using the BIOS;
the computing part comprises a computing part or a computer, and the computing unit can be each core of one computing part, or each computing part, GPU or TPU in the computer composed of multiple computing parts;
the operating power calculation module includes:
the formula is adopted:
calculating the current operating power of each core; g is a power threshold value, ps is the current core operating power, fs is an active frequency, fn is the rated frequency of each core of the calculation part, us is the current voltage of the calculation part, and Un is the rated voltage of the calculation part;
the kernel adjusting module comprises:
the FFT conversion unit is used for carrying out FFT conversion on the activity frequency of each core in two adjacent time slices to obtain the rising and falling trend of the activity frequency of each core;
the idle power calculation module is used for extracting an idle core with a descending activity frequency and calculating the idle power of the idle core;
and the power transfer module is used for adjusting the frequency multiplication coefficient of each core by using the BIOS and transferring the idle power of the idle core to the target core.
4. The system of claim 3, wherein the kernel adjustment module further comprises:
and the idle core adjusting unit is used for carrying out FFT (fast Fourier transform) on the active frequency of each core, judging whether the active frequency of the cores in the plurality of time slices is linearly increased or not, closing the corresponding idle core when the active frequency of the cores is not linearly increased, adjusting the frequency multiplication coefficient of each core by using the BIOS (basic input/output system), and transferring the idle power of the idle core to the target core.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211315622.4A CN115373858B (en) | 2022-10-26 | 2022-10-26 | Peak-shifting adjusting method and system based on constant power of calculating part |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211315622.4A CN115373858B (en) | 2022-10-26 | 2022-10-26 | Peak-shifting adjusting method and system based on constant power of calculating part |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115373858A CN115373858A (en) | 2022-11-22 |
CN115373858B true CN115373858B (en) | 2023-01-24 |
Family
ID=84073951
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211315622.4A Active CN115373858B (en) | 2022-10-26 | 2022-10-26 | Peak-shifting adjusting method and system based on constant power of calculating part |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115373858B (en) |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9811142B2 (en) * | 2014-09-29 | 2017-11-07 | Apple Inc. | Low energy processor for controlling operating states of a computer system |
US9652026B2 (en) * | 2014-12-21 | 2017-05-16 | Qualcomm Incorporated | System and method for peak dynamic power management in a portable computing device |
CN111247513B (en) * | 2017-04-17 | 2021-10-08 | 迪普西格有限公司 | Placement and scheduling of radio signal processing data stream operations |
-
2022
- 2022-10-26 CN CN202211315622.4A patent/CN115373858B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN115373858A (en) | 2022-11-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110928691B (en) | Traffic data-oriented edge collaborative computing unloading method | |
CN105868004B (en) | Scheduling method and scheduling device of service system based on cloud computing | |
US8547840B1 (en) | Bandwidth allocation of bursty signals | |
CN113672383A (en) | Cloud computing resource scheduling method, system, terminal and storage medium | |
CN113918240B (en) | Task unloading method and device | |
CN113286317A (en) | Task scheduling method based on wireless energy supply edge network | |
CN101819459B (en) | Heterogeneous object memory system-based power consumption control method | |
CN115373858B (en) | Peak-shifting adjusting method and system based on constant power of calculating part | |
CN111124639A (en) | Operation method and system of edge computing system and electronic equipment | |
CN111142647A (en) | Energy saving method and system for IT system | |
CN115759626A (en) | Multi-resource dynamic clustering method and system for improving K-means + | |
CN108241526B (en) | Adaptive control strategy based on mode prediction | |
CN110850957B (en) | Scheduling method for reducing system power consumption through dormancy in edge computing scene | |
CN102710766A (en) | Real-time access load evaluation-based cluster storage interface node configuration method | |
CN114020407A (en) | Container management cluster container group scheduling optimization method, device and equipment | |
CN117521905A (en) | Power generation equipment screening method and related device | |
CN115543609B (en) | Cloud computing virtual resource scheduling method based on clustering integration algorithm | |
CN110794949A (en) | Power consumption reduction method and system for automatically allocating computing resources based on component temperature | |
CN110825212A (en) | Energy-saving scheduling method and device and computer storage medium | |
CN114103707B (en) | Intelligent energy control method and system based on artificial intelligence and Internet of things | |
CN109376845B (en) | Dynamic adjustment method and dynamic adjustment coprocessor | |
CN101943943B (en) | Method for optimizing energy consumption of computation array based on similar resource aggregation | |
CN110809290A (en) | Method and system for optimizing utility of software-defined heterogeneous network with in-layer cooperation | |
CN116257360B (en) | Method and system for planning container group resources based on historical usage data | |
CN113673837A (en) | Carbon emission reduction method, system, terminal and storage medium for cloud data center |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |