CN107506286A - CPU and memory block automatically upper inserting method and system - Google Patents
CPU and memory block automatically upper inserting method and system Download PDFInfo
- Publication number
- CN107506286A CN107506286A CN201710882762.2A CN201710882762A CN107506286A CN 107506286 A CN107506286 A CN 107506286A CN 201710882762 A CN201710882762 A CN 201710882762A CN 107506286 A CN107506286 A CN 107506286A
- Authority
- CN
- China
- Prior art keywords
- cpu
- offline
- memory block
- memory
- lower line
- 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
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3024—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is a central processing unit [CPU]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3037—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is a memory, e.g. virtual memory, cache
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3065—Monitoring arrangements determined by the means or processing involved in reporting the monitored data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3089—Monitoring arrangements determined by the means or processing involved in sensing the monitored data, e.g. interfaces, connectors, sensors, probes, agents
-
- 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
Abstract
The invention discloses the automatically upper inserting method and system, methods described of CPU and memory block to pass through the CPU average loads of finger daemon real-time monitoring system and memory usage according to the upper offline strategy of selection;Reached the standard grade or during offline condition when meeting, the CPU of inquiry system and the presence bitmap of memory block, select the CPU or memory block for being capable of upper and lower line, carry out upper and lower line operation;After carrying out upper and lower line operation, corresponding CPU or memory block presence are updated to presence bitmap.The present invention is after the strategy pre-established is imported, for the service request of real-time change, whether server can need to carry out the upper offline of CPU and memory block with automatic decision, the automatic upper and lower line operation for carrying out CPU and memory block, reaction speed is fast, and decision-making is more preferably, improve the flexibility that high-end server uses, manpower management cost is also saved, reduces energy consumption under proper condition, reduces complete machine radiating.
Description
Technical field
The present invention relates to the operating system design technical field of high-end server, and in particular to CPU and memory block it is automatic
Upper inserting method and system.
Background technology
With the development of chip interconnection technique, the processor and internal memory of high-end server mounted inside are more and more,
Accordingly, the calculation processing power of high-end server is also stronger and stronger, and energy consumption is more and more.It is right in server internal
It is core or referred to as CPU in the management granularity of processor, and the management granularity of internal memory is the memory block of unified size.In high-end clothes
It is engaged in during the use of device, according to different business and different periods, sometimes and does not need all CPU and memory block same
Shi Yunzuo cans meet business demand.For the consideration of energy-conservation, included during marquis is to high-end server when appropriate some
CPU and memory block carry out offline operation, and offline CPU and memory block are then carried out operation of reaching the standard grade when needing again, this
Sample more flexibly and efficiently can use a high-end server, also more energy-conserving and environment-protective.
Prior art supports keeper to carry out manual CPU in the system administration webpage of high-end server, on memory block
Offline operation.CPU, memory block upper offline operation are carried out manually, and human cost is high, and reaction speed is slow, and keeper is often
Optimal operation, such as offline several cpu, several memory blocks can not be made.
The off-peak period of some business is being run, keeper carries out the CPU of server and the offline place of memory block manually
Reason can save energy consumption, but keeper can not often judge that offline how many CPU and memory block are not just influenceing service operation
Premise is issued to optimal energy-saving effect.Meanwhile if occurring the increased situation of service request amount suddenly, and keeper is not on duty
Position, then it can not in time be reached the standard grade by offline CPU and memory block, cause the performance of server can not all to play, business fortune
Row and Consumer's Experience can be affected.
The content of the invention
The technical problem to be solved in the present invention is:In view of the above-mentioned problems, the present invention provides the automatic of a kind of CPU and memory block
Upper inserting method and system.
The technical solution adopted in the present invention is:
CPU and memory block automatically upper inserting method, methods described content include:
According to the upper offline strategy of selection, pass through the CPU average loads of finger daemon real-time monitoring system and memory usage;
Reached the standard grade or during offline condition when meeting, the CPU of inquiry system and the presence bitmap of memory block, select can go up,
Offline CPU or memory block, carry out upper and lower line operation;
After carrying out upper and lower line operation, corresponding CPU or memory block presence are updated to presence bitmap.
Methods described content also includes:
By the upper and lower line operation note carried out in system journal, it is easy to occur analysis use during the system failure.
The upper offline tactful content includes:
If X is the threshold value of CPU average load, Y is the threshold value of memory usage;
When the CPU average loads of system are more than X, carry out CPU and reach the standard grade operation;
When the CPU average loads of system are less than X, CPU offline operation is carried out;
When the CPU average loads of system are equal to X, without CPU upper offline operation;
When the memory usage of system is more than Y, carries out internal memory and reach the standard grade operation;
When the memory usage of system is less than Y, the offline operation of internal memory is carried out;
When the memory usage of system is equal to Y, without the upper offline operation of internal memory.
In the upper offline strategy, by setting on CPU on offline decision-making time interval T1 and memory block during offline decision-making
Between be spaced T2, avoid excessively that frequently upper offline operation can influence the overall performance of server.
The threshold X of the average load of the CPU and the threshold value Y of memory usage set a fluctuation range, such as(X±
10%, Y ± 20%), instead of strategy(X,Y), further avoid excessively that frequently upper offline operation can influence the complete machine of server
Energy.
So, upper offline strategy can be expanded into(X ± m%, T1, Y ± n%, T2), wherein m, n are 1 ~ 99 nature
Number.
When CPU and memory block all meet offline condition, preferential offline CPU, offline internal memory can be suitably postponed, because whole
In terms of function consumption, in general CPU energy consumption accounting is more than the energy consumption accounting of internal memory, is conceived to saving energy consumption, it is of the invention from
The dynamic offline offline CPU of policy priority.
Automatic linear system system, the system up and down of CPU and memory block include:
Upper offline tactful import modul, CPU average loads and memory usage monitoring module, upper offline operation decision-making module, on
Offline operation decision-making module, wherein:
Upper offline tactful import modul is responsible for importing the upper offline strategy of selection;
CPU average loads and memory usage monitoring module are responsible for CPU average loads and the memory usage of real-time monitoring system;
Reached the standard grade or during offline condition when meeting, upper offline operation decision-making module is existed by the CPU of inquiry system and memory block
Wire state bitmap, the CPU or memory block for being capable of upper and lower line are selected, offline operation decision-making module carries out CPU or memory block in notice
Upper and lower line operation;
After upper offline operation decision-making module carries out upper and lower line operation, upper offline operation decision-making module is corresponding CPU or memory block
Presence is updated to presence bitmap.
The upper offline operation decision-making module by the upper and lower line operation note carried out in system journal, be easy to be
Analysis uses during failure of uniting.
Beneficial effects of the present invention are:
The present invention is after the strategy pre-established is imported, and for the service request of real-time change, server can be with automatic decision
Whether needing to carry out the upper offline of CPU and memory block, the automatic upper and lower line operation for carrying out CPU and memory block, reaction speed is fast,
Decision-making more preferably, improves the flexibility that high-end server uses, also saves manpower management cost, reduce under proper condition
Energy consumption, reduce complete machine radiating.Meanwhile system journal has been charged in upper offline operation every time, it is also convenient for follow-up once sending out
Positioning problems are carried out during raw mechanical disorder.
Brief description of the drawings
Fig. 1 is the inventive method upper offline tactful execution flow chart automatically;
Fig. 2 realizes system schematic for automatic upper inserting method.
Embodiment
According to Figure of description, with reference to embodiment, the present invention is further described:
Embodiment 1:
As shown in figure 1, the automatically upper inserting method of CPU and memory block, methods described content includes:
According to the upper offline strategy of selection, pass through the CPU average loads of finger daemon real-time monitoring system and memory usage;
Reached the standard grade or during offline condition when meeting, the CPU of inquiry system and the presence bitmap of memory block, select can go up,
Offline CPU or memory block, carry out upper and lower line operation;
After carrying out upper and lower line operation, corresponding CPU or memory block presence are updated to presence bitmap.
Methods described content also includes:
By the upper and lower line operation note carried out in system journal, it is easy to occur analysis use during the system failure.
The upper offline tactful content includes:
If X is the threshold value of CPU average load, Y is the threshold value of memory usage;
When the CPU average loads of system are more than X, carry out CPU and reach the standard grade operation;
When the CPU average loads of system are less than X, CPU offline operation is carried out;
When the CPU average loads of system are equal to X, without CPU upper offline operation;
When the memory usage of system is more than Y, carries out internal memory and reach the standard grade operation;
When the memory usage of system is less than Y, the offline operation of internal memory is carried out;
When the memory usage of system is equal to Y, without the upper offline operation of internal memory.
Embodiment 2
On the basis of embodiment 1, in upper offline strategy described in the present embodiment, by setting offline decision-making time interval on CPU
Offline decision-making time interval T2 on T1 and memory block, avoids excessively that frequently upper offline operation can influence the complete machine of server
Energy.
The threshold X of the average load of the CPU and the threshold value Y of memory usage set a fluctuation range, such as(X±
10%, Y ± 20%), instead of strategy(X,Y), further avoid excessively that frequently upper offline operation can influence the complete machine of server
Energy.
So, upper offline strategy can be expanded into(X ± m%, T1, Y ± n%, T2), wherein m, n are 1 ~ 99 nature
Number.
Embodiment 3
When CPU and memory block all meet offline condition, preferential offline CPU, offline internal memory can be suitably postponed, because in complete machine energy
In terms of consumption, in general CPU energy consumption accounting is more than the energy consumption accounting of internal memory, is conceived to saving energy consumption, it is of the invention it is automatic on
The offline offline CPU of policy priority.
Embodiment 4
As shown in Fig. 2 automatic linear system system, the system up and down of CPU and memory block include:
Upper offline tactful import modul, CPU average loads and memory usage monitoring module, upper offline operation decision-making module, on
Offline operation decision-making module, wherein:
Upper offline tactful import modul is responsible for importing the upper offline strategy of selection;
CPU average loads and memory usage monitoring module are responsible for CPU average loads and the memory usage of real-time monitoring system;
Reached the standard grade or during offline condition when meeting, upper offline operation decision-making module is existed by the CPU of inquiry system and memory block
Wire state bitmap, the CPU or memory block for being capable of upper and lower line are selected, offline operation decision-making module carries out CPU or memory block in notice
Upper and lower line operation;
After upper offline operation decision-making module carries out upper and lower line operation, upper offline operation decision-making module is corresponding CPU or memory block
Presence is updated to presence bitmap.
Embodiment 5
On the basis of embodiment 4, the upper offline upper and lower line operation note that operates decision-making module and will be carried out described in the present embodiment
In system journal, it is easy to occur analysis use during the system failure.
Embodiment is merely to illustrate the present invention, and not limitation of the present invention, the ordinary skill about technical field
Personnel, without departing from the spirit and scope of the present invention, it can also make a variety of changes and modification, thus it is all equivalent
Technical scheme fall within scope of the invention, scope of patent protection of the invention should be defined by the claims.
Claims (8)
1.CPU and memory block automatically upper inserting method, it is characterised in that methods described content includes:
According to the upper offline strategy of selection, pass through the CPU average loads of finger daemon real-time monitoring system and memory usage;
Reached the standard grade or during offline condition when meeting, the CPU of inquiry system and the presence bitmap of memory block, select can go up,
Offline CPU or memory block, carry out upper and lower line operation;
After carrying out upper and lower line operation, corresponding CPU or memory block presence are updated to presence bitmap.
2. CPU according to claim 1 and memory block automatically upper inserting method, it is characterised in that methods described content
Also include:
By the upper and lower line operation note carried out in system journal.
3. CPU according to claim 2 and memory block automatically upper inserting method, it is characterised in that the upper offline plan
Content slightly includes:
If X is the threshold value of CPU average load, Y is the threshold value of memory usage;
When the CPU average loads of system are more than X, carry out CPU and reach the standard grade operation;
When the CPU average loads of system are less than X, CPU offline operation is carried out;
When the CPU average loads of system are equal to X, without CPU upper offline operation;
When the memory usage of system is more than Y, carries out internal memory and reach the standard grade operation;
When the memory usage of system is less than Y, the offline operation of internal memory is carried out;
When the memory usage of system is equal to Y, without the upper offline operation of internal memory.
4. CPU according to claim 3 and memory block automatically upper inserting method, it is characterised in that the upper offline plan
In slightly, by setting on CPU offline decision-making time interval T2 on offline decision-making time interval T1 and memory block.
5. the automatically upper inserting method of CPU according to claim 4 and memory block, it is characterised in that methods described it is interior
Appearance also includes:
The threshold value Y of the threshold X of CPU average load and memory usage is set into a fluctuation range.
6. the automatically upper inserting method of the CPU and memory block according to any of the above-described claim, it is characterised in that work as CPU
When all meeting offline condition with memory block, preferential offline CPU.
7.CPU and memory block automatic linear system system up and down, it is characterised in that the system includes:
Upper offline tactful import modul, CPU average loads and memory usage monitoring module, upper offline operation decision-making module, on
Offline operation decision-making module, wherein:
Upper offline tactful import modul is responsible for importing the upper offline strategy of selection;
CPU average loads and memory usage monitoring module are responsible for CPU average loads and the memory usage of real-time monitoring system;
Reached the standard grade or during offline condition when meeting, upper offline operation decision-making module is existed by the CPU of inquiry system and memory block
Wire state bitmap, the CPU or memory block for being capable of upper and lower line are selected, offline operation decision-making module carries out CPU or memory block in notice
Upper and lower line operation;
After upper offline operation decision-making module carries out upper and lower line operation, upper offline operation decision-making module is corresponding CPU or memory block
Presence is updated to presence bitmap.
8. CPU according to claim 7 and memory block automatic linear system system up and down, it is characterised in that the upper offline behaviour
Module of making decision is by the upper and lower line operation note carried out in system journal.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710882762.2A CN107506286A (en) | 2017-09-26 | 2017-09-26 | CPU and memory block automatically upper inserting method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710882762.2A CN107506286A (en) | 2017-09-26 | 2017-09-26 | CPU and memory block automatically upper inserting method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107506286A true CN107506286A (en) | 2017-12-22 |
Family
ID=60699402
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710882762.2A Pending CN107506286A (en) | 2017-09-26 | 2017-09-26 | CPU and memory block automatically upper inserting method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107506286A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109408242A (en) * | 2018-11-13 | 2019-03-01 | 郑州云海信息技术有限公司 | Inserting method and device on a kind of server resource |
CN109471724A (en) * | 2018-10-23 | 2019-03-15 | 郑州云海信息技术有限公司 | A kind of method and system of switch operating amount of memory |
CN110851330A (en) * | 2019-11-13 | 2020-02-28 | 重庆秦嵩科技有限公司 | CPU on-line monitoring method |
WO2021101584A1 (en) * | 2019-11-18 | 2021-05-27 | Sandisk Technologies Llc | Core controller architecture |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101673223A (en) * | 2009-10-22 | 2010-03-17 | 同济大学 | Thread dispatching implementation method based on on-chip multiprocessor |
CN102222014A (en) * | 2011-06-16 | 2011-10-19 | 华中科技大学 | Dynamic memory management system based on memory hot plug for virtual machine |
CN102520782A (en) * | 2011-12-15 | 2012-06-27 | 江苏中科梦兰电子科技有限公司 | Power supply management method based on automatic adjustment of processor nuclear number |
CN102567107A (en) * | 2011-10-31 | 2012-07-11 | 广东电网公司电力科学研究院 | Highly-concurrent real-time memory resource management and scheduling method |
CN104503838A (en) * | 2014-11-23 | 2015-04-08 | 华中科技大学 | Method for scheduling virtual CPU (Central Processing Unit) |
CN105955809A (en) * | 2016-04-25 | 2016-09-21 | 深圳市万普拉斯科技有限公司 | Thread scheduling method and system |
JP2017107486A (en) * | 2015-12-11 | 2017-06-15 | 富士通株式会社 | Processing resource control program, processing resource controller, and processing resource control method |
-
2017
- 2017-09-26 CN CN201710882762.2A patent/CN107506286A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101673223A (en) * | 2009-10-22 | 2010-03-17 | 同济大学 | Thread dispatching implementation method based on on-chip multiprocessor |
CN102222014A (en) * | 2011-06-16 | 2011-10-19 | 华中科技大学 | Dynamic memory management system based on memory hot plug for virtual machine |
CN102567107A (en) * | 2011-10-31 | 2012-07-11 | 广东电网公司电力科学研究院 | Highly-concurrent real-time memory resource management and scheduling method |
CN102520782A (en) * | 2011-12-15 | 2012-06-27 | 江苏中科梦兰电子科技有限公司 | Power supply management method based on automatic adjustment of processor nuclear number |
CN104503838A (en) * | 2014-11-23 | 2015-04-08 | 华中科技大学 | Method for scheduling virtual CPU (Central Processing Unit) |
JP2017107486A (en) * | 2015-12-11 | 2017-06-15 | 富士通株式会社 | Processing resource control program, processing resource controller, and processing resource control method |
CN105955809A (en) * | 2016-04-25 | 2016-09-21 | 深圳市万普拉斯科技有限公司 | Thread scheduling method and system |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109471724A (en) * | 2018-10-23 | 2019-03-15 | 郑州云海信息技术有限公司 | A kind of method and system of switch operating amount of memory |
CN109408242A (en) * | 2018-11-13 | 2019-03-01 | 郑州云海信息技术有限公司 | Inserting method and device on a kind of server resource |
CN109408242B (en) * | 2018-11-13 | 2020-08-04 | 郑州云海信息技术有限公司 | Server resource online and offline method and device |
CN110851330A (en) * | 2019-11-13 | 2020-02-28 | 重庆秦嵩科技有限公司 | CPU on-line monitoring method |
CN110851330B (en) * | 2019-11-13 | 2023-04-21 | 重庆秦嵩科技有限公司 | CPU on-line monitoring method |
WO2021101584A1 (en) * | 2019-11-18 | 2021-05-27 | Sandisk Technologies Llc | Core controller architecture |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107506286A (en) | CPU and memory block automatically upper inserting method and system | |
CA2741088C (en) | Methods of achieving cognizant power management | |
CN112243285B (en) | Base station energy saving method and device | |
CN103475790B (en) | A kind of intelligent mobile terminal power consumption management | |
WO2016041468A1 (en) | Wakeup method, device and terminal | |
CN103176842B (en) | background process management method and mobile terminal thereof | |
CN102711139B (en) | A kind of data flow statistic method and terminal | |
CN106059835B (en) | A kind of High-reliability Control method of low energy consumption computer set group node | |
CN102521662A (en) | Consumption charging method in cloud computing environment | |
CN110138688A (en) | Dynamic adjusts method, apparatus, equipment and the readable storage medium storing program for executing of business interface | |
CN104184884B (en) | A kind of control device and method, mobile terminal of mobile terminal power consumption | |
CN107844187A (en) | Power consumption management method, device and electronic equipment | |
CN105430030A (en) | OSG-based parallel extendable application server | |
CN105955444A (en) | Aligned wakeup method and apparatus | |
CN108550930A (en) | Bridging management device, bridge system and bridging method based on retired power battery pack | |
CN106681475A (en) | Method for reducing power consumption of mobile terminal and mobile terminal | |
CN104796673B (en) | A kind of cloud video monitoring system task cut-in method towards energy optimization | |
CN102831016B (en) | Physical machine recycle method of cloud computing and device thereof | |
CN102546999B (en) | Method, control device and system for reducing device power consumption based on business model | |
CN104345858B (en) | Power consumption management method and device, the server of server hardware device | |
CN106371481A (en) | Terminal energy-saving management method and apparatus, and terminal | |
CN106200866A (en) | A kind of application freezing method and mobile terminal | |
CN115225685B (en) | Method and device for controlling equipment power consumption | |
CN104111892A (en) | Internal memory dynamic distribution method and system | |
CN208368671U (en) | Bridging management device and bridge system based on retired power battery pack |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171222 |