CN107832204A - A kind of server CPU energy consumption Forecasting Methodologies based on MDC - Google Patents
A kind of server CPU energy consumption Forecasting Methodologies based on MDC Download PDFInfo
- Publication number
- CN107832204A CN107832204A CN201711224544.6A CN201711224544A CN107832204A CN 107832204 A CN107832204 A CN 107832204A CN 201711224544 A CN201711224544 A CN 201711224544A CN 107832204 A CN107832204 A CN 107832204A
- Authority
- CN
- China
- Prior art keywords
- prediction
- energy consumption
- mdc
- cpu
- calculating
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3466—Performance evaluation by tracing or monitoring
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3452—Performance evaluation by statistical analysis
-
- 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)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention discloses a kind of server CPU energy consumption Forecasting Methodologies based on MDC, it is related to data center field, provide a kind of prediction algorithm of CPU idleness, define the time span of prediction, algorithm is automatically adjusted according to current predictive result, the relation of the change of CPU energy consumptions and CPU temperature rises change is defined, and prediction algorithm and MDC modular data center dynamic environment monitoring systems are linked so that energy consumption monitoring, prediction and the dynamic environment monitoring system of server are organically blended；Meet MDC dynamic environment monitoring systems for safe class it is high the needs of, lift the supporting ability of dynamic environment monitoring system；Predicted by energy consumption, improve the efficiency of operation of data center, the change for CPU energy consumptions and temperature provides forewarning function, solves the problems, such as that current only monitor is not predicted well, improves the O＆M safe class of data center.
Description
Technical Field
The invention relates to the field of data centers, in particular to a server CPU energy consumption prediction method based on MDC.
Background
With the continuous development of the scale of the cloud data center and the gradual maturity of the artificial intelligence technology, the demand for predicting the energy consumption of the CPU of the data center server is urgent. It can be seen that energy consumption monitoring in a cloud data center is a very practical and urgent function. Monitoring energy consumption data is the basis for scheduling based on energy consumption. For example, the energy consumption monitoring senses that a certain server in the data center is currently in a low energy consumption state, and senses that the server is to maintain the low energy consumption state according to energy consumption prediction, so that the service on the server can be migrated to other servers according to an energy consumption scheduling strategy, and the server is in a closed or dormant state, so that the purpose of reducing the energy consumption can be achieved.
The Modular Data Center (MDC) is a new generation of Data Center deployment form based on cloud computing, and in order to cope with the trend of server development such as cloud computing, virtualization, centralization, and densification, the MDC adopts a modular design concept to reduce the coupling of infrastructure to the machine room environment to the greatest extent. The modular data center MDC integrates subsystems such as power supply and distribution, refrigeration, cabinet, airflow containment, comprehensive wiring, dynamic environment monitoring and the like, improves the overall operation efficiency of the data center, and realizes quick deployment, elastic expansion and green energy conservation.
The single mechanism power density of the modular data center MDC is large and can reach more than 10kw, so that the modular data center MDC has very important significance for monitoring and predicting the energy consumption of the server CPU. At present, in the aspect of energy consumption monitoring, a modular data center MDC mainly monitors hardware, and can only monitor energy consumption monitoring data without a prediction function.
Disclosure of Invention
Aiming at the requirements and the defects of the prior art development, the invention provides a server CPU energy consumption prediction method based on MDC.
The invention relates to a server CPU energy consumption prediction method based on MDC, which solves the technical problems and adopts the following technical scheme: the server CPU energy consumption prediction method based on the MDC provides a CPU idle rate prediction algorithm, defines the prediction time span, automatically adjusts the algorithm according to the current prediction result, defines the relation between the CPU energy consumption change and the CPU temperature rise change, and links the prediction algorithm with an MDC modular data center power environment monitoring system, so that the energy consumption monitoring and prediction of the server and the power environment monitoring system are organically integrated; the specific implementation process comprises the following steps:
step one, a CPU idle rate prediction algorithm is given;
step two, automatically adjusting a prediction algorithm according to a current prediction result;
and step three, the prediction algorithm is linked with a dynamic environment monitoring system of the MDC modular data center.
Further, in the first step, a prediction algorithm for the CPU idle rate is given:
and obtaining two predicted values of the CPU idle rate in a period of time by adopting a weighted moving average method and an exponential smoothing method, fitting the two predicted values, inputting the time span needing prediction, and outputting the predicted value of the energy consumption.
Further, in the second step, the prediction algorithm is automatically adjusted according to the current prediction result, and the specific implementation flow includes:
step 1, forming a historical monitoring record sequence set;
step 2, determining a sample set of a historical monitoring record sequence;
step 3, calculating the weight of the monitoring record;
step 4, calculating a prediction result of a weighted moving average method;
step 5, calculating a predicted value of an exponential smoothing method;
step 6, calculating the deviation degree of the price predicted by the weighted moving average method;
step 7, calculating the deviation degree of the prediction result of the exponential smoothing method;
and 8, calculating the gain of the weighted moving average method, and calculating a final predicted value.
Further, in step 1, the reading time span is 2T pred The set of history sequences A = (v) 0 ,v 1 ,v 2 ,…,v 2Tpred-1 )；
In the step 2, the current time point is taken as a reference starting point, and the modular length not less than T is selected from the set A pred And the historical record sequence with the minimum variance forms a sample set of the historical monitoring record sequence
Further, in step 3, after the sample set H is determined, the calculation of each monitoring record is performed according to the following formulaWeight w i
Further, in step 4, after the weight of each monitoring record is obtained through calculation, a weighted average value is calculated by using the following formula
And taking the weighted average as a prediction result, wherein the prediction result of the weighted moving average method is as follows:
further, in the step 5, a predicted value of the exponential smoothing method is calculatedWherein, the first and the second end of the pipe are connected with each other,the latest predicted value of the exponential smoothing method is shown, and the calculation formula is as follows:
further, in the step 6, the prediction result of the weighted moving average method is calculatedDegree of deviation ofWhereinIs the result of the prediction in the last period,is the most recent record in the history sequence.
Further, in the step 7, the prediction result of the exponential smoothing method is calculatedDegree of deviation ofWhereinIs the result of the prediction of the last period,is the most recent record in the history sequence.
Further, in the step 8, the step of,
if it isThe weighted moving average gain amount is calculated according to the following formulaGetThenCalculating a final predicted value
If it isCalculating a moving weighted average according to the following formulaAmount of gain by lawGet theThen theCalculating a final predicted value
Compared with the prior art, the MDC-based server CPU energy consumption prediction method has the beneficial effects that: the invention provides a CPU idle rate prediction algorithm, which is linked with an MDC modular data center power environment monitoring system, meets the requirement of the MDC power environment monitoring system on high safety level, and improves the service support capability of the power environment monitoring system; by energy consumption prediction, the operation efficiency of the data center is improved, an early warning effect is provided for the change of CPU energy consumption and temperature, the problem that the data center is only monitored but not predicted at present is well solved, and the operation and maintenance safety level of the data center is improved.
Detailed Description
In order to make the technical solutions, technical problems to be solved, and technical effects of the present invention more clearly apparent, the technical solutions of the present invention are described below in detail with reference to specific embodiments. All embodiments that can be obtained by a person skilled in the art without making any inventive step on the basis of the embodiments of the present invention are within the scope of protection of the present invention.
The power consumption of the physical server can be accurately expressed as a linear relation of the CPU utilization rate of the physical server, and the power consumption of the physical server in an idle state is about 70% of the power consumption of the physical server in a full utilization state.
The functional relationship between the power consumption of the physical server and the CPU idle rate is as follows:
P(u idle )＝0.7*P max +(1-0.7)*P max *(1-u idle )
thus, using the above formula, a prediction of CPU energy consumption can be translated into a prediction of CPU idleness.
Example 1:
the embodiment provides a server CPU energy consumption prediction method based on MDC, and the implementation process comprises the following steps:
step one, a CPU idle rate prediction algorithm is given;
step two, automatically adjusting a prediction algorithm according to a current prediction result;
and step three, the prediction algorithm is linked with a power environment monitoring system of the MDC modular data center, so that the requirement of the power environment monitoring system on high safety level is met, and the service support capability of the power environment monitoring system of the MDC is improved.
The embodiment of the server CPU energy consumption prediction method based on the MDC provides a CPU idle rate prediction algorithm, the prediction time span is defined, the algorithm is automatically adjusted according to the current prediction result, the relation between the CPU energy consumption change and the CPU temperature rise change is defined, and the prediction algorithm is linked with an MDC modular data center dynamic environment monitoring system, so that the energy consumption monitoring, prediction and dynamic environment monitoring system of the server are organically integrated, and an efficient, stable and reliable global monitoring management method of the MDC dynamic environment monitoring system is constructed.
Example 2:
in the server CPU energy consumption prediction method based on MDC provided in this embodiment, a detailed technical solution of step 1 is added on the basis of the server CPU energy consumption prediction method based on MDC in embodiment 1, and two types of trend analysis algorithms, namely a weighted moving average method and a weighted exponential smoothing method, are adopted to cooperate with each other, so that the self-correcting capability of the prediction algorithm is improved, and further, the CPU energy consumption prediction work is better performed.
For the prediction of the CPU idle rate, only the long-term trend of the CPU idle rate can be considered, and the existing long-term trend prediction algorithm has single processing on the variation trend, so that the prediction algorithm cannot well conform to the actual situation, and the self-correction capability is poor.
In the first step of this embodiment, a prediction algorithm for the CPU idle rate is given, and the specific contents include:
obtaining predicted values of CPU idle rates in two periods of time by adopting a weighted moving average method and an exponential smoothing methodAndthen fitting the two predicted values, and inputting the time span T to be predicted pred Output the predicted value of energy consumptionThe self-correcting capability can be improved, and then prediction work can be better carried out.
Example 3:
in the method for predicting the energy consumption of the server CPU based on the MDC provided in this embodiment, a specific technical solution of the second step is added on the basis of the method for predicting the energy consumption of the server CPU based on the MDC in embodiment 2, which is helpful for improving the operation and maintenance security level of the MDC modular data center and providing an early warning effect for the change of the PCU energy consumption and the temperature.
And step two, automatically adjusting the prediction algorithm according to the current prediction result, wherein the specific implementation flow is as follows:
step 1, forming a historical monitoring record sequence set;
specifically, the read time span is 2T pred The set of history sequences A = (v) 0 ,v 1 ,v 2 ,…,v 2Tpred-1 )；
Step 2, determining a sample set of a historical monitoring record sequence;
specifically, the current time point is taken as a reference starting point, and the modular length not less than T is selected from the set A pred And the history record sequence with the minimum variance forms a set
T pred <n≤2T pred Wherein, in the step (A),indicating the earliest record in the selected sequence of history records,representing the latest record in the history sequence, wherein the set H has a modular length of n;
step 3, calculating the weight of the monitoring record;
specifically, after the sample set H is determined, the weight w of each monitoring record is calculated according to the following formula i
Step 4, calculating a prediction result of a weighted moving average method;
specifically, after the weight of each monitoring record is calculated, the weighted average is calculated by using the following formula
In the weighted moving average method, if the weighted average is used as the prediction result, the prediction result of the weighted moving average method is:
step 5, calculating a predicted value of an exponential smoothing method;
specifically, the predicted value of the exponential smoothing method is calculatedWherein the content of the first and second substances,the latest prediction value which is finished by the exponential smoothing method is represented, and the calculation formula is as follows:
step 6, calculating the deviation degree of the price predicted by the weighted moving average method;
specifically, the prediction result of the weighted moving average method is calculatedDegree of deviation ofWhereinIs the result of the prediction of the last period,is the most recent record in the history sequence;
step 7, calculating the deviation degree of the prediction result of the exponential smoothing method;
specifically, the prediction result of the exponential smoothing method is calculatedDegree of deviation ofWhereinIs the result of the prediction of the last period,is the most recent record in the history sequence;
step 8, calculating the gain of the weighted moving average method, and calculating the final predicted value;
in particular, ifThe weighted moving average gain amount is calculated according to the following formulaGetThen theCalculating a final predicted value
If it isCalculating the gain of the moving weighted average method according to the following formulaGet theThenCalculating a final predicted value
The prediction of the energy consumption of the CPU of the server of the MDC is reflected by the temperature change rate of the CPU, the temperature of the CPU is higher when the energy consumption of the CPU is larger, and the change of the internal environment temperature monitored by the MDC dynamic environment monitoring system is much slower than the temperature change of the CPU, so that the effect of early warning cannot be achieved. In order to solve the above problems of the modular data center MDC, the method for predicting the energy consumption of the server CPU based on the MDC according to the embodiment can monitor the change and the change rate of the temperature of the CPU at the same time, and provide an early warning service for an extreme scene.
The principle and embodiments of the present invention are described in detail by using specific examples, which are only used to help understanding the core technical content of the present invention, and are not used to limit the protection scope of the present invention, and the technical solution of the present invention is not limited to the specific embodiments described above. Based on the above embodiments of the present invention, those skilled in the art should make any improvements and modifications to the present invention without departing from the principle of the present invention, and therefore, the present invention should fall into the protection scope of the present invention.
Claims (10)
1. A server CPU energy consumption prediction method based on MDC is characterized in that a CPU idle rate prediction algorithm is given, a prediction time span is defined, the algorithm is automatically adjusted according to a current prediction result, the relation between CPU energy consumption change and CPU temperature rise change is defined, and the prediction algorithm is linked with an MDC modular data center power environment monitoring system, so that the energy consumption monitoring and prediction of a server are organically integrated with the power environment monitoring system; the specific implementation process comprises the following steps:
step one, a prediction algorithm of the CPU idle rate is given;
step two, automatically adjusting a prediction algorithm according to a current prediction result;
and step three, the prediction algorithm is linked with a dynamic environment monitoring system of the MDC modular data center.
2. The MDC-based server CPU energy consumption prediction method according to claim 1, wherein the first step, a prediction algorithm giving CPU idle rate:
and obtaining two predicted values of the CPU idle rate in a period of time by adopting a weighted moving average method and an exponential smoothing method, fitting the two predicted values, inputting the time span needing prediction, and outputting the predicted value of the energy consumption.
3. The method for predicting the CPU energy consumption of the MDC-based server according to claim 2, wherein in the second step, the prediction algorithm is automatically adjusted according to the current prediction result, and the specific implementation process comprises the following steps:
step 1, forming a historical monitoring record sequence set;
step 2, determining a sample set of a historical monitoring record sequence;
step 3, calculating the weight of the monitoring record;
step 4, calculating a prediction result of a weighted moving average method;
step 5, calculating a predicted value of an exponential smoothing method;
step 6, calculating the deviation degree of the price predicted by the weighted moving average method;
step 7, calculating the deviation degree of the prediction result of the exponential smoothing method;
and 8, calculating the gain of the weighted moving average method, and calculating a final predicted value.
4. The MDC-based server CPU energy consumption prediction method according to claim 3, wherein in the step 1, the reading time span isConstitutes a history sequence set a = (v) 0 ,v 1 ,v 2 ,…,v 2Tpred-1 )；
Step 2, selecting the modular length not less than T in the set A by taking the current time point as a reference starting point pred And the historical record sequence with the minimum variance forms a sample set of the historical monitoring record sequence
5. The MDC-based server CPU energy consumption prediction method according to claim 4, wherein in the step 3, after the sample set H is determined, the weight w of each monitoring record is calculated according to the following formula i
6. The MDC-based server CPU energy consumption prediction method as claimed in claim 4, wherein in the step 4, after the weight value of each monitoring record is obtained through calculation, the weighted average value is calculated by using the following formula
And taking the weighted average as a prediction result, wherein the prediction result of the weighted moving average method is as follows:
7. the MDC-based server CPU energy consumption prediction method according to claim 6, wherein the step 5 of calculating the predicted value of the exponential smoothing methodWherein the content of the first and second substances,the latest predicted value of the exponential smoothing method is shown, and the calculation formula is as follows:
8. the MDC-based server CPU energy consumption prediction method as claimed in claim 7, wherein the step 6 is to calculate the prediction result by weighted moving average methodDegree of deviation ofWhereinIs the result of the prediction in the last period,is the most recent record in the history sequence.
9. The MDC-based server CPU energy consumption prediction method as claimed in claim 8, wherein the step 7 of calculating the prediction result of exponential smoothing methodDegree of deviation ofWhereinIs the result of the prediction in the last period,is the most recent record in the history sequence.
10. The MDC-based server CPU energy consumption prediction method of claim 9, wherein the step 8,
if it isThe weighted moving average gain amount is calculated according to the following formulaGet theThen theCalculating a final predicted value
If it isCalculating the gain of the moving weighted average method according to the following formulaB = (I)Then theCalculating a final predicted value
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711224544.6A CN107832204A (en) | 2017-11-29 | 2017-11-29 | A kind of server CPU energy consumption Forecasting Methodologies based on MDC |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711224544.6A CN107832204A (en) | 2017-11-29 | 2017-11-29 | A kind of server CPU energy consumption Forecasting Methodologies based on MDC |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107832204A true CN107832204A (en) | 2018-03-23 |
Family
ID=61646651
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711224544.6A Pending CN107832204A (en) | 2017-11-29 | 2017-11-29 | A kind of server CPU energy consumption Forecasting Methodologies based on MDC |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107832204A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112019581A (en) * | 2019-05-30 | 2020-12-01 | 华为技术有限公司 | Method and device for scheduling task processing entities |
| CN113934615A (en) * | 2021-12-15 | 2022-01-14 | 山东中创软件商用中间件股份有限公司 | Data monitoring method, device and equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103401699A (en) * | 2013-07-18 | 2013-11-20 | 深圳先进技术研究院 | Cloud data center security monitoring early warning system and method |
| CN103853918A (en) * | 2014-02-21 | 2014-06-11 | 南京邮电大学 | Cloud computing server dispatching method based on idle time prediction |
| KR20150049499A (en) * | 2013-10-30 | 2015-05-08 | 국민대학교산학협력단 | Energy saving apparatus and method using resource usage pattern of software |
| CN107239849A (en) * | 2017-04-17 | 2017-10-10 | 西安电子科技大学 | One kind is based on improved exponential smoothing gray model Methods of electric load forecasting |
-
2017
- 2017-11-29 CN CN201711224544.6A patent/CN107832204A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103401699A (en) * | 2013-07-18 | 2013-11-20 | 深圳先进技术研究院 | Cloud data center security monitoring early warning system and method |
| KR20150049499A (en) * | 2013-10-30 | 2015-05-08 | 국민대학교산학협력단 | Energy saving apparatus and method using resource usage pattern of software |
| CN103853918A (en) * | 2014-02-21 | 2014-06-11 | 南京邮电大学 | Cloud computing server dispatching method based on idle time prediction |
| CN107239849A (en) * | 2017-04-17 | 2017-10-10 | 西安电子科技大学 | One kind is based on improved exponential smoothing gray model Methods of electric load forecasting |
Non-Patent Citations (1)
| Title |
|---|
| 曲红梅 等: ""组合模型对恶性肿瘤死亡率拟合度评价及预测方法的研究"", 《中华流行病学杂志》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112019581A (en) * | 2019-05-30 | 2020-12-01 | 华为技术有限公司 | Method and device for scheduling task processing entities |
| CN112019581B (en) * | 2019-05-30 | 2022-02-25 | 华为技术有限公司 | Method and device for scheduling task processing entities |
| CN113934615A (en) * | 2021-12-15 | 2022-01-14 | 山东中创软件商用中间件股份有限公司 | Data monitoring method, device and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Fang et al. | Thermal-aware energy management of an HPC data center via two-time-scale control | |
| WO2018137402A1 (en) | Cloud data centre energy-saving scheduling implementation method based on rolling grey prediction model | |
| CN107036238B (en) | Intelligent energy-saving control method for dynamically predicting external air and load | |
| CN109028446A (en) | A kind of refrigerating plant room control method based on equipment overall performance model | |
| WO2021082478A1 (en) | Energy consumption prediction method and device for air conditioning system | |
| WO2019169706A1 (en) | Demand prediction method, demand control method, and system | |
| WO2014084941A1 (en) | Analytics for optimizing usage of cooling subsystems | |
| CN103423189A (en) | Server fan power consumption measuring method | |
| CN115309603A (en) | Data center energy consumption prediction optimization method, system, medium and computing device | |
| CN107832204A (en) | A kind of server CPU energy consumption Forecasting Methodologies based on MDC | |
| CN111191851A (en) | Data center energy efficiency optimization method based on knowledge graph | |
| CN114063545A (en) | Data center energy consumption control system and method fusing edge calculation and controller | |
| JP2014180134A (en) | Power management system, and power management method | |
| CN114221353B (en) | Power grid regulation method, device, equipment, storage medium and computer program product | |
| CN113326585B (en) | Energy efficiency abnormality early warning method and device for gas boiler and computer equipment | |
| CN114909945A (en) | Energy-saving intelligent control method and device for cooling tower, electronic device and medium | |
| CN112711229B (en) | Intelligent optimization energy-saving system based on multi-correlation factor energy consumption prediction | |
| CN104680010B (en) | A kind of Steam Turbine steady-state operation data screening method | |
| US10599204B1 (en) | Performance efficiency monitoring system | |
| CN117273284A (en) | Abnormal data monitoring system for enterprise electricity balance | |
| Kumar et al. | Power usage efficiency (PUE) optimization with counterpointing machine learning techniques for data center temperatures | |
| CN115437876A (en) | Data center management method and device, electronic equipment and storage medium | |
| Xue et al. | A novel method of minimizing power consumption for existing chiller plant | |
| CN106094523B (en) | Based on efficiency and flow index area and maximum parallel operation system optimization method | |
| Pittino et al. | A scalable framework for online power modelling of high-performance computing nodes in production |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180323 |
|
| RJ01 | Rejection of invention patent application after publication |