CN117785481B - Data center computing resource allocation management system - Google Patents
Data center computing resource allocation management system Download PDFInfo
- Publication number
- CN117785481B CN117785481B CN202410199317.6A CN202410199317A CN117785481B CN 117785481 B CN117785481 B CN 117785481B CN 202410199317 A CN202410199317 A CN 202410199317A CN 117785481 B CN117785481 B CN 117785481B
- Authority
- CN
- China
- Prior art keywords
- computing node
- computing
- calculation
- time
- node equipment
- 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
- 238000007726 management method Methods 0.000 title claims abstract description 18
- 238000013468 resource allocation Methods 0.000 title claims abstract description 11
- 238000004364 calculation method Methods 0.000 claims abstract description 73
- 238000004458 analytical method Methods 0.000 claims abstract description 17
- 238000011156 evaluation Methods 0.000 claims abstract description 7
- 238000000034 method Methods 0.000 claims description 32
- 230000004044 response Effects 0.000 claims description 16
- 230000006978 adaptation Effects 0.000 claims description 14
- 230000008859 change Effects 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 241001334134 Rugopharynx epsilon Species 0.000 claims 1
- 230000002159 abnormal effect Effects 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000005457 optimization Methods 0.000 description 5
- 230000003044 adaptive effect Effects 0.000 description 3
- 230000036962 time dependent Effects 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Landscapes
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a data center computing resource allocation management system, which belongs to the technical field of computing resource management and comprises: the data calculation management terminal acquires a calculation task submitted by a user and temporarily stores the calculation task; the sequence evaluation module is used for determining the calculation priority of the calculation task submitted by the user; the computing node analysis module is used for determining the load condition coefficient and the running condition coefficient of each computing node device; and the task distribution module is used for comprehensively analyzing the computing node equipment according to the acquired load condition coefficient and the running condition coefficient, so that the computing task is distributed to the corresponding computing node equipment. The invention comprehensively analyzes the load condition coefficient of the equipment and the operation condition coefficient of the equipment, and can more clearly know the condition of each computing node equipment, so that the computing tasks are preferentially distributed to the computing node equipment with better operation condition and lower load, and the overall computing efficiency and the resource utilization rate are improved.
Description
Technical Field
The invention belongs to the field of computing resource management, and particularly relates to a data center computing resource allocation management system.
Background
A data center is a location for centralized processing, storage, transmission, management, and exchange of data, and is comprised of a set of facilities including servers, storage devices, network devices, power supplies, cooling systems, etc. for supporting large-scale data processing and application operations. Data centers are commonly used to store and manage large amounts of data and to provide data services such as data backup, data recovery, data synchronization, and the like. Data centers are also considered a strategic resource because they are globally coordinated network of specific devices that communicate, accelerate, display, calculate, store data information over the Internet, a network infrastructure.
When computing resource allocation is performed in a data center, computing tasks are generally and uniformly allocated to each computing node device to perform data computation, but because load conditions of different node devices may be different, and each node device has different operation conditions due to reasons of service life, self configuration and the like, if uniform allocation is performed, it is easy for part of node devices to be already computed, and part of devices are still in a computing state, time is delayed, and overall computing efficiency of the computing center is affected.
Disclosure of Invention
The present invention is directed to a data center computing resource allocation management system, which is used for solving the above-mentioned problems in the prior art.
The aim of the invention can be achieved by the following technical scheme:
a data center computing resource allocation management system, the management system comprising:
The data calculation management end is used for acquiring calculation tasks submitted by users and temporarily storing the calculation tasks;
The sequence evaluation module is used for determining the calculation priority of the calculation tasks submitted by the user;
the computing node analysis module is used for determining the load condition coefficient and the running condition coefficient of each computing node device;
and the task distribution module is used for comprehensively analyzing the computing node equipment according to the acquired load condition coefficient and the running condition coefficient so as to distribute the computing task to the corresponding computing node equipment.
Further, the method for determining the calculation priority by the order evaluation module comprises the following steps:
Determining a computing priority in a first-come-last order according to the time of submitting the computing task by the user;
When multiple users submit computing tasks at the same time, priorities are determined from simple to complex orders according to the complexity of the computing tasks.
Further, the complexity determining method comprises the following steps:
By the formula Solving a complex coefficient D of a calculation task submitted by each user;
wherein m is the type of data calculation type in each calculation task, and i is E ，/>Calculating the complexity of the type for the i-th data,/>And G is the size of the calculation task for the calculation parameters contained in the ith data type.
Further, the method for obtaining the load condition coefficient of each computing node device by the computing node analysis module comprises the following steps:
obtaining CPU utilization rate of each computing node device Memory occupancy/>And disk usage rateSimultaneously acquiring the number S and the waiting time T of tasks to be processed under the computing node;
By the formula Solving the load condition coefficient/>, of each computing node device；
Wherein,/>For a preset proportionality coefficient,/>Is a preset standard reference time.
Further, the method for obtaining the running condition coefficients of each computing node device by the computing node analysis module comprises the following steps:
when the computing node equipment works, a curve of the operation rate in the fatting time along with the time is obtained And a temperature change over time curve/>；
By the formulaSolving an operation condition coefficient K of the computing node equipment;
wherein, ，/>And a standard temperature change time-dependent curve preset for the system.
Further, the method for performing calculation task allocation by the task allocation module comprises the following steps:
By the formula Calculate the adaptation degree/>Wherein P is an adaptive function preset by the system;
According to the acquired adaptation degree And sequencing the computing node devices from high to low, and distributing the computing tasks to the computing node device with the highest adaptation degree.
Further, the computing node analysis module is further configured to determine, according to the operation condition coefficient, a potential fault condition of the computing node device, where the determining method includes:
the obtained operation condition coefficient K of the computing node equipment and a preset threshold interval of the system Comparison is performed:
When (when) When the computing node equipment fails, the computing node equipment is indicated;
When (when) When the method is used, further judgment is carried out;
When (when) And when the computing node equipment is normal.
Further, the further judging method comprises the following steps:
Acquiring network traffic rate magnitude of computing node device per second in t time ；
Simultaneously acquiring response time of each calculation of the computing node equipment in t time；
By the formulaSolving for risk value/>, of computing node equipment；
Wherein q is the total seconds in the time of t and，/>For the mean rate, n is the total number of times the response is calculated over the time of t, and/>，/>For response time at jth calculation,/>Standard response time preset for the system in the jth calculation,/>/>Is a weight coefficient;
Will obtain a risk value Risk threshold preset with System/>Comparing;
When (when) When the computing node equipment fails, the computing node equipment is indicated;
Otherwise, the computing node device is indicated to be normal.
The invention has the beneficial effects that:
The invention comprehensively analyzes the load condition coefficient of the equipment and the operation condition coefficient of the equipment, and can more clearly know the condition of each computing node equipment, so that the computing tasks are preferentially distributed to the computing node equipment with better operation condition and lower load, and the overall computing efficiency and the resource utilization rate are improved.
According to the method and the device, whether the equipment is abnormal in fault or not can be determined by comprehensively analyzing the acquired operation condition coefficient of the computing node equipment and combining the network fluctuation condition or the optimization degree condition of computing software in the equipment, the potential fault condition of the computing node equipment can be judged, and the abnormal equipment can be timely maintained, so that the computing efficiency of the computing equipment is ensured.
Of course, it is not necessary for any one product to practice the invention to achieve all of the advantages set forth above at the same time.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a system block diagram of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In one embodiment, a data center computing resource allocation management system is disclosed, as shown in FIG. 1, comprising:
the data calculation management end is used for acquiring a calculation task submitted by a user and temporarily storing the calculation task;
The sequence evaluation module is used for determining the calculation priority of the calculation task submitted by the user;
The computing node analysis module is used for determining the load condition coefficient and the running condition coefficient of each computing node device;
And the task distribution module is used for comprehensively analyzing the computing node equipment according to the acquired load condition coefficient and the running condition coefficient so as to distribute the computing task to the corresponding computing node equipment.
According to the technical scheme, the order evaluation module performs calculation priority evaluation on submitted tasks, so that reasonable allocation is performed, waiting time of an ordering queue can be reduced, meanwhile, load condition coefficients and operation condition coefficients of all the computing node devices are acquired through the computing node analysis module, so that computing devices with the best states in all the computing node devices are found out according to the conditions of the load condition coefficients and the operation condition coefficients, the current tasks are issued to the current computing devices, the computing tasks can be allocated according to the condition of the computing tasks, computing resources can be utilized more reasonably, the time of data computation is shortened, and the computing efficiency of the whole computing center is guaranteed.
As one embodiment of the invention, the method for determining the calculation priority by the sequence evaluating module comprises the following steps:
Determining a computing priority in a first-come-last order according to the time of submitting the computing task by the user;
when a plurality of users submit calculation tasks at the same time, determining priority from simple to complex according to the complexity of the calculation tasks;
The complexity determining method comprises the following steps: by the formula Solving a complex coefficient D of a calculation task submitted by each user;
wherein m is the type of data calculation type in each calculation task, and i is E ，/>Calculating the complexity of the type for the i-th data,/>And G is the size of the calculation task for the calculation parameters contained in the ith data type.
Through the technical scheme, the embodiment provides a specific method for determining the computing priority of the computing tasks, firstly, each computing task is ordered according to the time when the user submits the computing tasks and the sequence from beginning to end, so that the priority is determined, each computing task can be reasonably and orderly planned, data confusion is avoided, and subsequent resource allocation is facilitated; when a plurality of users submit calculation tasks at the same time, priorities are determined from a simple to complex sequence according to the complexity of the calculation tasks, so that the same submitted calculation tasks are processed firstly, then the complex data are processed, the overall waiting time can be reduced macroscopically, the overall processing efficiency can be improved, if the complex calculation tasks are processed firstly, more resources (such as memory and processing capacity) are needed, the system resources are possibly tense, and the calculation resources can be more effectively utilized by the simple tasks. And the simple calculation task is processed firstly, so that a foundation can be provided for the subsequent complex calculation task, when the complex calculation task comprises the expansion of the simple calculation task, the processed simple task can be used as a starting point, and the development time and complexity are reduced, so that the processing efficiency is improved. And the complexity determining method is realized by a formulaThe complexity coefficient D of the calculation task submitted by each user is obtained, the complexity degree is determined according to the size of the complexity coefficient, and the formula shows that when the scale G of the whole calculation task is smaller, or the complexity degree/>, of each data calculation type contained in the calculation taskThe lower the calculation parameter/>, theThe fewer the complexity that accounts for the overall computational task, the smaller the complexity factor D; through the operation, comprehensive analysis can be performed according to the scale of the task to be calculated, the complexity of each data calculation type contained in the calculation task and the contained calculation parameters, so that the complexity of the calculation task is known clearly, and the subsequent task allocation is facilitated.
In the above technical solution, the size G of the computing task can be formulated according to the total number of bytes, the number of rows, the number of records, and the like of the data in the computing task at the time of submission, and the complexity of each data computing typeAnd the processing time of the related data in the big data is determined, and the description is not repeated.
As one implementation mode of the invention, the method for acquiring the load condition coefficients of each computing node device by the computing node analysis module comprises the following steps:
obtaining CPU utilization rate of each computing node device Memory occupancy/>And disk usage rateSimultaneously acquiring the number S and the waiting time T of tasks to be processed under the computing node;
By the formula Solving the load condition coefficient/>, of each computing node device；
Wherein,/>For a preset proportionality coefficient,/>Is a preset standard reference time.
With the above technical solution, the present embodiment provides a specific method for obtaining the load status factor of each computing node device, and in general, the load status of the computing device is critical to the computing efficiency, because an excessive load may cause a decrease in computing performance or a resource exhaustion, and the load status is generally related to the usage of the CPU, the memory, the disk, etc., so as to obtain the CPU utilization of each computing node deviceMemory occupancy/>And disk usage/>Then through the formula/>Determining load condition coefficients of each computing node deviceAs can be seen from the formula, when the CPU utilization is higher, the memory occupancy is higher or the disk utilization is higher, the load of the computing device is larger, namely/>The greater the value of (2); also, if the number of tasks to be processed in the computing device is greater, or the waiting time is longer, the more tasks the computing device needs to process in the future, the greater the load is, that is/>The larger the value of (2), the more so by the formula/>The comprehensive analysis is carried out, so that the load condition coefficient of each computing node device can be obtained, and when Y is smaller, the lower the load is indicated; through the operation, the load condition of each computing node device can be clearly known, so that computing tasks are distributed to computing node devices with lighter loads according to the load condition, and the overall computing efficiency and the resource utilization rate are improved.
In the technical proposal, the preset proportion coefficient/>Preset standard reference time/>Are all empirically developed based on historical data and are not described in any great detail herein.
As one implementation mode of the invention, the method for acquiring the operation condition coefficient of each computing node device by the computing node analysis module comprises the following steps:
when the computing node equipment works, a curve of the operation rate in the fatting time along with the time is obtained And a temperature change over time curve/>；
By the formulaSolving an operation condition coefficient K of the computing node equipment;
wherein, ，/>And a standard temperature change time-dependent curve preset for the system.
Through the technical scheme, the embodiment provides a specific method for acquiring the operation condition coefficient of each computing node device, and generally, the faster the device is calculated, the smaller the temperature change is, which indicates that the better the operation condition of the device is; thus, when the computing node equipment works, the curve of the operation speed in the fatting time along with the time is obtainedAnd a temperature change over time curve/>Then through the formula/>Solving the operation condition coefficient K of the computing node equipment, and finding out when/>, from the formulaThe larger the value of (2) is, the faster the computing node device computes the rate in the fatt time, and when/>The smaller the value, the smaller the temperature change value of the computing node device during the father time is, the better the device operation condition is, thus when/>The larger the value of (2), the better the operation condition of the computing node equipment is; through the operation, the operation condition of each computing node device can be clearly known, so that computing tasks are distributed to the computing node devices with lighter loads according to the operation condition, and the overall computing efficiency and the resource utilization rate are improved.
In the above technical scheme, t is a preset period of time,And a standard temperature change time-dependent change curve preset by the system is drawn according to historical data of equipment operation.
As one implementation mode of the invention, the method for carrying out calculation task allocation by the task allocation module comprises the following steps:
By the formula Calculate the adaptation degree/>Wherein P is an adaptive function preset by the system;
According to the acquired adaptation degree And sequencing the computing node devices from high to low, and distributing the computing tasks to the computing node device with the highest adaptation degree.
Through the above technical solution, the present embodiment provides a specific method for performing computing task allocation by the task allocation module, which includes performing continuous simulation training according to historical data to obtain a preset adaptive function P, and performing a formulaCalculate the adaptation degree/>Thereby obtaining the adaptation degree of each computing node device, and then according to the obtained adaptation degree/>Sequencing all the computing node devices from high to low, and distributing computing tasks to the computing node device with the highest adaptation degree; through the operation, the load condition coefficient of the equipment and the operation condition coefficient of the equipment are comprehensively analyzed, so that the condition of each computing node equipment can be more clearly known, and the computing tasks are preferentially distributed to the computing node equipment with better operation condition and lower load, so that the overall computing efficiency and the resource utilization rate are improved.
As an embodiment of the present invention, the computing node analysis module is further configured to determine, according to the operation condition coefficient, a potential fault condition of the computing node device, where the determining method includes:
the obtained operation condition coefficient K of the computing node equipment and a preset threshold interval of the system Comparison is performed:
When (when) When the computing node equipment fails, the computing node equipment is indicated;
When (when) When the method is used, further judgment is carried out;
When (when) When the computing node equipment is normal;
The method for further judging is as follows: acquiring network traffic rate magnitude of computing node device per second in t time ；
Simultaneously acquiring response time of each calculation of the computing node equipment in t time；
By the formulaSolving for risk value/>, of computing node equipment；
Wherein q is the total seconds in the time of t and，/>For the mean rate, n is the total number of times the response is calculated over the time of t, and/>，/>For response time at jth calculation,/>Standard response time preset for the system in the jth calculation,/>/>Is a weight coefficient;
Will obtain a risk value Risk threshold preset with System/>Comparing;
When (when) When the computing node equipment fails, the computing node equipment is indicated;
Otherwise, the computing node device is indicated to be normal.
Through the above technical solution, the present embodiment provides a specific method for determining a potential fault condition of a computing node device, which indicates that there is a high probability of a fault when the operating condition of the computing node device is poor, and in order to ensure the computing efficiency, repair and maintenance are required for the device, so that the obtained operating condition coefficient K of the computing node device and a threshold interval preset by a system are obtainedComparison, when/>When the operation condition of the computing node equipment is poor, the computing node equipment is required to be repaired and maintained; when/>When the operation state of the computing node equipment is better, the equipment works normally; when/>When the equipment is in fault, further judgment is needed, and the further judgment method is realized through a formulaSolving for risk value/>, of computing node equipment; Since the network fluctuation or the abnormality of the computing software in the equipment is possible when the equipment fails, and the equipment itself is not necessarily failed, the formula/>The network fluctuation condition of the equipment is obtained, and the smaller the value of the equipment is, the smaller the network fluctuation is, the more stable the network is, and the lower the abnormal probability of the network is; then through the formulaThe software response condition is obtained, and the smaller the difference value between the response time in each calculation and the preset standard response time is, the higher the optimization degree of the calculation software is, and the lower the abnormality probability of the software is; finally through the formulaComprehensive analysis is carried out to obtain the risk value/>, of the computing node equipmentAnd will obtain risk values/>Risk threshold preset with System/>Comparing whenAt the moment, the network is stable or the optimization degree of the software is high, the risk of faults of the computing node equipment is high, and maintenance is neededAnd when the network fluctuation is unstable or the software is low in optimization degree, the running condition of the equipment is abnormal, and the equipment is not faulty, so that the computing node equipment is normal. Through the operation, whether the equipment is abnormal in fault or not can be determined according to the acquired operation condition coefficient of the computing node equipment and by combining network fluctuation conditions or the optimization degree conditions of computing software in the equipment, the potential fault condition of the computing node equipment can be judged, and the abnormal equipment is timely maintained, so that the computing efficiency of the computing equipment is ensured.
In the above technical solution, the system presets a threshold intervalRisk threshold/>Standard response time/>, preset by system, at each calculationCan be formulated according to the relevant historical data, and the weight coefficient/>AndIt may be empirically determined and will not be described further herein.
The foregoing is merely illustrative and explanatory of the principles of the invention, as various modifications and additions may be made to the specific embodiments described, or similar thereto, by those skilled in the art, without departing from the principles of the invention or beyond the scope of the appended claims.
Claims (3)
1. A data center computing resource allocation management system, the management system comprising:
the data calculation management terminal is used for acquiring calculation tasks submitted by users and temporarily storing the calculation tasks;
The sequence evaluation module is used for determining the calculation priority of the calculation tasks submitted by the user;
the computing node analysis module is used for determining the load condition coefficient and the running condition coefficient of each computing node device;
The task distribution module is used for comprehensively analyzing the computing node equipment according to the acquired load condition coefficient and the running condition coefficient so as to distribute the computing task to the corresponding computing node equipment;
The method for acquiring the load condition coefficients of each computing node device by the computing node analysis module comprises the following steps:
the method comprises the steps of obtaining CPU utilization rate L cpu, memory occupancy rate L memory and disk utilization rate L disk of each computing node device, and obtaining the number S of tasks to be processed and waiting time T under the computing node;
By the formula Solving a load condition coefficient Y of each computing node device;
Wherein, alpha 1 and alpha 2 are preset proportionality coefficients, and T 0 is preset standard reference time;
The method for acquiring the running condition coefficients of each computing node device by the computing node analysis module comprises the following steps:
When the computing node equipment works, acquiring a calculation rate time-varying curve V (t) and a temperature time-varying curve W (t) in delta t time;
By the formula Solving an operation condition coefficient K of the computing node equipment;
Wherein Δt=t 2-t1,W0 (t) is a standard temperature change curve preset by the system along with time;
the method for distributing the calculation tasks by the task distribution module comprises the following steps: by the formula Solving an adaptation degree R, wherein P is an adaptation function preset by a system;
According to the obtained adaptation degree R, sequencing all the computing node devices from high to low, and distributing computing tasks to the computing node device with the highest adaptation degree;
The computing node analysis module is also used for judging potential fault conditions of the computing node equipment according to the running condition coefficients, and the judging method comprises the following steps:
Comparing the obtained operation condition coefficient K of the computing node equipment with a preset threshold interval [ K x,Ky ] of the system:
when K epsilon (0, K x), indicating that the computing node equipment has faults;
When K is E [ K x,Ky ], further judgment is carried out;
When K is E (K y, ++ infinity), the computing node equipment is normal;
the further judging method comprises the following steps: acquiring the network flow rate VL r of each second of the computing node equipment in delta t time;
Simultaneously acquiring response time TC j of the computing node equipment in each computation within delta t time;
By the formula Solving a risk value risk of the computing node equipment;
where q is the total seconds in Δt time, and r.epsilon. (1, q), For the mean rate, n is the total number of calculated responses in Δt time, and j ε (1, n), TC j is the response time at the jth calculation,/>The standard response time preset for the system in the j-th calculation is beta 1 and beta 2 which are weight coefficients;
Comparing the obtained risk value risk with a risk threshold value risk th preset by the system;
When risk < risk th, indicating that the computing node equipment has a fault;
Otherwise, the computing node device is indicated to be normal.
2. The system for managing allocation of computing resources in a data center of claim 1, wherein said order assessment module determines the computing priority by:
Determining a computing priority in a first-come-last order according to the time of submitting the computing task by the user;
When multiple users submit computing tasks at the same time, priorities are determined from simple to complex orders according to the complexity of the computing tasks.
3. The data center computing resource allocation management system of claim 2, wherein the complexity determination method is as follows:
By the formula Solving a complex coefficient D of a calculation task submitted by each user;
Wherein M is the type of the data calculation type in each calculation task, i is e (1, M), F i is the complexity of the ith data calculation type, M i is the calculation parameter contained in the ith data type, and G is the size of the calculation task.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410199317.6A CN117785481B (en) | 2024-02-23 | 2024-02-23 | Data center computing resource allocation management system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410199317.6A CN117785481B (en) | 2024-02-23 | 2024-02-23 | Data center computing resource allocation management system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117785481A CN117785481A (en) | 2024-03-29 |
| CN117785481B true CN117785481B (en) | 2024-05-24 |
Family
ID=90389191
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410199317.6A Active CN117785481B (en) | 2024-02-23 | 2024-02-23 | Data center computing resource allocation management system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117785481B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101656706B1 (en) * | 2015-04-02 | 2016-09-22 | 두산중공업 주식회사 | Job distribution system in high-performance computing environment |
| CN111459617A (en) * | 2020-04-03 | 2020-07-28 | 南方电网科学研究院有限责任公司 | Containerized application automatic allocation optimization system and method based on cloud platform |
| KR20220006490A (en) * | 2021-12-29 | 2022-01-17 | 케이웨어 (주) | Hybrid cloud resource allocation method for workload dynamic resource placement and optimization performance management |
| CN117221069A (en) * | 2023-09-14 | 2023-12-12 | 研华科技(中国)有限公司 | Management method and device based on micro-server architecture |
| CN117436763A (en) * | 2023-11-30 | 2024-01-23 | 广州墨斗信息科技有限公司 | Method and system for realizing complete process fine management of building labor based on digitization |
| CN117573307A (en) * | 2023-11-13 | 2024-02-20 | 纬创软件(武汉)有限公司 | Method and system for overall management of multiple tasks in cloud environment |
-
2024
- 2024-02-23 CN CN202410199317.6A patent/CN117785481B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101656706B1 (en) * | 2015-04-02 | 2016-09-22 | 두산중공업 주식회사 | Job distribution system in high-performance computing environment |
| CN111459617A (en) * | 2020-04-03 | 2020-07-28 | 南方电网科学研究院有限责任公司 | Containerized application automatic allocation optimization system and method based on cloud platform |
| KR20220006490A (en) * | 2021-12-29 | 2022-01-17 | 케이웨어 (주) | Hybrid cloud resource allocation method for workload dynamic resource placement and optimization performance management |
| CN117221069A (en) * | 2023-09-14 | 2023-12-12 | 研华科技(中国)有限公司 | Management method and device based on micro-server architecture |
| CN117573307A (en) * | 2023-11-13 | 2024-02-20 | 纬创软件(武汉)有限公司 | Method and system for overall management of multiple tasks in cloud environment |
| CN117436763A (en) * | 2023-11-30 | 2024-01-23 | 广州墨斗信息科技有限公司 | Method and system for realizing complete process fine management of building labor based on digitization |
Non-Patent Citations (1)
| Title |
|---|
| 开源云计算资源调度策略优化研究;邱亚 等;《计算机技术与发展》;20230630;第33卷(第6期);第8-15页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117785481A (en) | 2024-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11497144B2 (en) | Optimized thermal control of data center | |
| Das et al. | Optimal preventive maintenance in a production inventory system | |
| US7467291B1 (en) | System and method for calibrating headroom margin | |
| CN106899660A (en) | Cloud data center energy-saving distribution implementation method based on trundle gray forecast model | |
| US20100228861A1 (en) | Environmental and computing cost reduction with improved reliability in workload assignment to distributed computing nodes | |
| CN105718364A (en) | Dynamic assessment method for ability of computation resource in cloud computing platform | |
| JP2009176304A (en) | Power control system | |
| US20210255899A1 (en) | Method for Establishing System Resource Prediction and Resource Management Model Through Multi-layer Correlations | |
| CN105872061B (en) | A kind of server set group managing means, apparatus and system | |
| CN109117269A (en) | A kind of distributed system dispatching method of virtual machine, device and readable storage medium storing program for executing | |
| CN118093326B (en) | Cloud rendering resource monitoring method and system based on computing power scheduling | |
| CN112433807A (en) | Airflow perception type virtual machine scheduling method oriented to data center global energy consumption optimization | |
| US10599204B1 (en) | Performance efficiency monitoring system | |
| CN117785481B (en) | Data center computing resource allocation management system | |
| CN110957724B (en) | Power distribution network operation mode and grid structure evaluation method | |
| CN118193188A (en) | Dynamic load balancing system and method based on virtualization platform | |
| CN117555913A (en) | Object data updating method and device based on third party platform | |
| CN117370138A (en) | High capacity distributed storage system | |
| CN111935952A (en) | Large machine room energy consumption regulation and control method and device | |
| CN115314500B (en) | Dynamic load balancing method based on improved TOPSIS model | |
| CN115469974A (en) | Kubernetes Pod elastic expansion and contraction capacity method under resource limited state | |
| CN116527677A (en) | Cloud disk thermomigration method of distributed disaster recovery system based on gateway node load | |
| CN111210060B (en) | Method for predicting temperature of machine room during working days | |
| CN107241752B (en) | A kind of the YARN dispatching method and system of sensing network flow | |
| CN114745282B (en) | Resource allocation model prediction method and device and electronic equipment |
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 |