CN108021453A - A kind of computing resource optimization method, device and server cluster - Google Patents
A kind of computing resource optimization method, device and server cluster Download PDFInfo
- Publication number
- CN108021453A CN108021453A CN201711405545.0A CN201711405545A CN108021453A CN 108021453 A CN108021453 A CN 108021453A CN 201711405545 A CN201711405545 A CN 201711405545A CN 108021453 A CN108021453 A CN 108021453A
- Authority
- CN
- China
- Prior art keywords
- execution unit
- computing resource
- calculating task
- calculate node
- resource
- 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
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/5022—Workload threshold
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
Abstract
The present invention provides a kind of computing resource optimization method, device and server cluster, the described method includes：Obtain the use information that the first calculating task handled in calculate node at least one calculating task takes computing resource；Based on the use information, the configuration information that the calculate node handles first calculating task is adjusted.Through the above it can be seen that, above-mentioned computing resource optimization method, pass through the use information to taking computing resource by obtaining the first calculating task handled in calculate node at least one calculating task, and the use information is based on, adjust the configuration information that the calculate node handles first calculating task.The computing resource optimization method is enabled to take the actual use information of computing resource according to calculating task, dynamic adjustment calculate node handles the configuration information of the calculating task, so as to more fully utilize whole computing resource, the overall computational efficiency of computing resource is improved.
Description
Technical field
The present embodiments relate to processor technical field, more particularly to a kind of computing resource optimization method, device and clothes
Business device cluster.
Background technology
The cluster that big data calculating platform is made of multiple calculate nodes (such as computer or server).It is existing big
Data calculating platform used hardware resource in the task of processing, such as the CPU of each computing unit in each calculate node
Sum of resource, memory source and computing unit etc., is pre-configured with or using cluster default configuration in the task of submission
's.
But in actual use, since the diversity of waiting task, such as waiting task may stress to count
Calculate, it is also possible to stress input output band width etc., it is easy to the cpu resource of each calculate node, memory source, net in cluster occur
The unbalanced situation that network resource and disk resource use, causes the problem of computing resource inclination occur.
For example, the utilization rate of the cpu resource of some waiting task is very high, but the utilization rate of disk resource is very low, this is just
It is very possible to make CPU occur using bottleneck.For another example some relatively simple waiting tasks, under default configuration, Wu Fachong
Divide the computing resource using whole cluster, such as the computing unit of acquiescence is total larger, and the cpu resource of each computing unit,
The utilization rate of memory source is all very low, this has resulted in the significant wastage of computing resource.Existing method can not solve these
Problem.
The content of the invention
An embodiment of the present invention provides a kind of computing resource optimization method, device and server cluster, can dynamically adjust
The configuration information of current calculate node processing calculating task is performed, so as to optimize the profit of calculate node and cluster to computing resource
With, reduce computing resource waste, improve calculate efficiency.
In order to solve the above technical problem, the present invention provides a kind of computing resource optimization method, including：
Obtain the use that the first calculating task handled in calculate node at least one calculating task takes computing resource
Information；
Based on the use information, the configuration information that the calculate node handles first calculating task is adjusted.
Preferably, wherein, the computing resource includes at least one of：
Processor resource；
Memory source；
Internet resources；
Hard disk resources.
Preferably, wherein, first calculating task is handled by least one execution unit, the execution unit configuration
There are processing core and memory source, the adjustment calculate node handles the configuration information of first calculating task, including：
Adjust the configuration of at least one execution unit.
Preferably, the described method includes at least one of：
Adjust the total quantity of the execution unit；With
Adjust the processing core quantity of the execution unit；With
Adjust the memory source size of the execution unit.
Preferably, described be based on the use information, adjust the calculate node and handle first calculating task
Configuration information, including at least one of：
When processor resource utilization rate is less than first threshold, the processing core quantity of the execution unit is reduced；With
When the processor resource utilization rate is higher than second threshold, increase the processing core quantity of the execution unit；
With
When memory resource utilization is less than three threshold values, the memory source size of the execution unit is reduced；With
When the memory source utilization rate is higher than four threshold values, increase the memory source size of the execution unit；With
When the ratio in high usage state is higher than five threshold values in the computing resource, increase the execution unit
Total quantity；With
When the ratio in poor efficiency state is higher than six threshold values in the computing resource, the execution unit is reduced
Total quantity；With
When the Internet resources and/or hard disk resources are unsatisfactory for requiring, matching somebody with somebody at least one execution unit is adjusted
Put.
Present invention also offers a kind of computing resource to optimize device, including：
Acquiring unit, it is used to obtain the first calculating task handled in calculate node at least one calculating task and takes
The use information of computing resource；
Adjustment unit, it is used to be based on the use information, adjusts the calculate node and handles first calculating task
Configuration information.
Preferably, wherein, the computing resource includes at least one of：
Processor resource；
Memory source；
Internet resources；
Hard disk resources.
Preferably, further include：
Execution unit, first calculating task are handled by least one execution unit, the execution unit configuration
There are processing core and memory source, the adjustment calculate node handles the configuration information of first calculating task, including：
Adjust the configuration of at least one execution unit.
Preferably, the adjustment unit is used at least one of：
Adjust the total quantity of the execution unit；With
Adjust the processing core quantity of the execution unit；With
Adjust the memory source size of the execution unit.
Present invention also offers a kind of server cluster, including：
Processor；
Memory, wherein, executable instruction is stored with the memory, wherein, it is described in the executable instruction
When processor performs so that the processor proceeds as follows：
Obtain the use that the first calculating task handled in calculate node at least one calculating task takes computing resource
Information；
Based on the use information, the configuration information that the calculate node handles first calculating task is adjusted.
Based on disclosed above, the beneficial effect of the embodiment of the present invention is：By to being handled by obtaining in calculate node
The first calculating task at least one calculating task takes the use information of computing resource, and is based on the use information, adjusts
The whole calculate node handles the configuration information of first calculating task.The computing resource optimization method enable according to meter
Calculation task takes the actual use information of computing resource, and dynamic adjustment calculate node handles the configuration information of the calculating task,
So as to more fully utilize whole computing resource, the overall computational efficiency of computing resource is improved.
Brief description of the drawings
Fig. 1 is the flow chart of the computing resource optimization method in the embodiment of the present invention；
Fig. 2 is the frame diagram of the computing resource optimization device in the embodiment of the present invention；
Fig. 3 is the frame diagram of the server cluster in the embodiment of the present invention.
Embodiment
To make those skilled in the art be better understood from technical scheme, below in conjunction with the accompanying drawings and specific embodiment party
Formula elaborates the present invention.
Herein with reference to the various schemes and feature of the attached drawing description present invention.
It is of the invention by the description to the preferred form of the embodiment that is given as non-limiting examples with reference to the accompanying drawings
These and other characteristic will become apparent.
It is also understood that although with reference to some instantiations, invention has been described, but people in the art
Member realize with can determine the present invention many other equivalents, they have feature as claimed in claim and therefore all
In the protection domain limited whereby.
When read in conjunction with the accompanying drawings, in view of described further below, above and other aspect of the invention, feature and advantage will become
It is more readily apparent.
Hereinafter with reference to the specific embodiment of the attached drawing description present invention；It will be appreciated, however, that the embodiment invented is only
The example of the present invention, it can use various ways to implement.Function and structure that is known and/or repeating is not described in detail with basis
The operation of the history of user, distinguishes real intention, avoids unnecessary or unnecessary details from make it that the present invention is smudgy.Cause
This, the specific structural and functional details invented herein are not intended to restriction, but as just the base of claim
Plinth and representative basis are used to instruct those skilled in the art diversely to use this hair with substantially any appropriate detailed construction
It is bright.
This specification can be used phrase " in one embodiment ", " in another embodiment ", " in another embodiment
In " or " in other embodiments ", it may refer to one or more of identical or different embodiment according to the present invention.
The present invention provides a kind of computing resource optimization method, including：
Obtain the use that the first calculating task handled in calculate node at least one calculating task takes computing resource
Information；
Based on the use information, the configuration information that the calculate node handles first calculating task is adjusted.
Through the above as can be seen that above-mentioned computing resource optimization method, by by obtaining in calculate node
The use information that the first calculating task at least one calculating task takes computing resource is handled, and letter is used based on described
Breath, adjusts the configuration information that the calculate node handles first calculating task.Enable the computing resource optimization method
The actual use information of computing resource, the dynamic configuration for adjusting calculate node and handling the calculating task are taken according to calculating task
Information, so as to more fully utilize whole computing resource, improves the overall computational efficiency of computing resource.
In order to more fully understand above-mentioned technical proposal, below in conjunction with Figure of description and specific embodiment pair
The specific flow of above-mentioned computing resource optimization method is illustrated.
As shown in Figure 1, Fig. 1 is the flow chart of the computing resource optimization method in the embodiment of the present invention, the described method includes
Following steps：
Step 101：Obtain the first calculating task handled in calculate node at least one calculating task and take calculating money
The use information in source；
Wherein, calculate node is used to handle at least one calculating task, and appoints handling at least one calculate
Computing resource can be utilized during business, calculate node can be server.
Calculating task, is task to be treated, needs to take computing resource and is completed by calculate node.
During actual use, which can be the comparison of data, the processing of image or download of video resource etc..
Computing resource, be for during calculating task is handled the required resource used, the computing resource can be
4 resource of processor, memory source, Internet resources, hard disk resources etc..
Use information, is the information using the specific service condition of computing resource.
Below using calculate node as server, calculating task be image is handled, computing resource is 4 resource of processor
Exemplified by illustrate.
At this time, step 101 is to obtain server when handling image, the processor which takes
The information of the service condition of 4 resources, for example, the use information obtained can be 4 resource utilization of processor for 30%, 50% or
80% etc..
In one embodiment, the computing resource includes 4 resource of processor, memory source, Internet resources and hard disk money
At least one of source.
It is below to carry out processing to image to illustrate using calculate node as server, calculating task.
At this time, step 101 is to obtain server when handling image, the processor which takes
In the use information of 4 resources, the use information of memory source, the use information of the use information of Internet resources and hard disk resources
It is a kind of or several, for example, the utilization rate that the use information obtained can be 4 resource of processor is 30%, the profit of memory source
With rate be 20%, the utilization rate of Internet resources is 10% and the utilization rate of hard disk resources is 2% etc..
Step 102：Based on the use information, adjust the calculate node processing first calculating task matches somebody with somebody confidence
Breath.
Wherein, configuration information is the calculate node required correlation used when first calculating task is handled
Information.
Further, in the present embodiment, first calculating task is handled by least one execution unit 3, described to hold
Row unit 3 is configured with processing core and memory source, and the adjustment calculate node handles matching somebody with somebody for first calculating task
Confidence ceases, including：
Adjust the configuration of at least one execution unit 3.
Wherein, the execution unit 3 is the unit that the calculate node is used to perform processing calculating task, certainly, in reality
During the use of border, when calculate node handles calculating task, the quantity of used execution unit 3 can be one
Can also be multiple.
Meanwhile in the present embodiment, the computing resource optimization method includes at least one of：
Adjust the total quantity of the execution unit 3；With the processing core quantity for adjusting the execution unit 3；With adjustment institute
State the memory source size of execution unit 3.
Specifically, it is described to be based on the use information, adjust the calculate node and handle matching somebody with somebody for first calculating task
Confidence ceases, including at least one of：
When 4 resource utilization of processor is less than first threshold, the processing core quantity of the execution unit 3 is reduced；With
When 4 resource utilization of processor is higher than second threshold, increase the process cores calculation of the execution unit 3
Amount；With
When memory resource utilization is less than three threshold values, the memory source size of the execution unit 3 is reduced；With
When the memory source utilization rate is higher than four threshold values, increase the memory source size of the execution unit 3；With
When the ratio in high usage state is higher than five threshold values in the computing resource, increase the execution unit
3 total quantity, wherein, the high usage state is exactly that 4 resource utilization of processor is higher than second threshold and/or memory source
Utilization rate is in high load capacity situation higher than resources such as the 4th threshold values；With
When the ratio in poor efficiency state is higher than six threshold values in the computing resource, the execution unit is reduced
3 total quantity, wherein, the poor efficiency is exactly that 4 resource utilization of processor is utilized less than first threshold and/or memory source
Rate is in underload situation less than resources such as the 3rd threshold values；With
When the Internet resources and/or hard disk resources are unsatisfactory for requiring, matching somebody with somebody at least one execution unit 3 is adjusted
Put.
Illustrated below by taking calculating task is a kind of algorithm of calculating as an example.
First, calculate node is obtained when handling the algorithm, the use information of 4 resource of processor of occupancy, at this time, due to
The algorithm comparison is simple, and during the treatment, the utilization rate for handling 4 resource of processor that the algorithm uses is only 1%, and pre-
The first threshold first set is 5%, that is, 4 resource utilization of processor actually used is less than first threshold, and then, reduce and perform
The processing core quantity of unit 3, on the one hand so that the utilization rate of 4 resource of processor rises above the normal value model of first threshold
Enclose, on the other hand, the processing core for the execution unit 3 for adjusting out can be used for handling other calculating tasks, so as to improve
The overall utilization rate of resource.
And when the algorithm comparison is complicated, during the treatment, handle 4 resource of processor that the algorithm uses
Utilization rate is up to 95%, and pre-set second threshold is 80%, that is, 4 resource utilization of processor actually used is higher than the
Two threshold values, and then, the adjustment to execution unit 3 is to increase the quantity of its processing core, so that the utilization of 4 resource of processor
Rate falls the range of normal value to second threshold after rise, meanwhile, improve the treatment effeciency for handling the algorithm.Certainly, should in increase
While the quantity of the processing core of execution unit 3, the size of the memory source of the execution unit 3 can also be increased, so that
Obtaining execution unit 3 can efficiently be handled when the algorithm is handled with rational configuration.
Illustrated below by taking calculating task is that a file is handled as an example.
Equally, calculate node is first obtained when handling this document, the use information of the memory source of occupancy, at this document
Manage relatively easy, in processing procedure, the utilization rate for handling the memory source of this document is only 1%, and the pre-set 3rd
Threshold value is 3%, that is, the utilization rate of the memory source actually used is less than the 3rd threshold value, and then, just reduce execution unit 3 memory
The size of resource, on the one hand so that the utilization rate of memory source rises above the range of normal value of first threshold, on the other hand,
The memory source of the execution unit 3 of adjustment out can be used for handling other calculating tasks, so as to improve the overall profit of resource
With rate.
And when this document processing is relative complex, during the treatment, the profit for the memory source that processing this document uses
95% is up to rate, and pre-set 4th threshold value is 85%, that is, the memory source utilization rate actually used is higher than the 4th threshold
Value, and then, the adjustment to execution unit 3 is to increase the size of its memory source, so that the utilization rate of memory source is fallen after rise
Range of normal value to the 4th threshold value, meanwhile, improve the treatment effeciency for handling this document.Certainly, the execution list is being increased
While the memory source size of member 3, the quantity of the processing core of the execution unit 3 can also be increased, so that performing list
Member 3 can efficiently be handled when this document is handled with rational configuration.
Or illustrated so that calculating task is that a file is handled as an example.
When this document is relative complex, and the computing resource of the processing this document obtained is in high usage state, such as, locate
The utilization rate for 4 resource of processor that reason this document uses is up to 90% and/or handles the utilization for the memory source that this document uses
Rate is up to 92%, naturally it is also possible to is that the utilization rate for handling the Internet resources that this document uses is up to 90% or processing this article
The utilization rate for the hard disk resources that part uses is up to 90%, i.e., the one or more in above-mentioned computing resource are more than one default the
Five threshold values, and pre-set 5th threshold value is 85%, at this point it is possible to increase the total quantity of execution unit 3, so that above-mentioned place
One or more of utilization rates in reason device 4 resource, memory source, Internet resources or hard disk resources fall after rise to the 5th threshold value with
Under normal range (NR), meanwhile, improve processing this document treatment effeciency.
And when this document is relatively easy, and the computing resource of processing this document obtained is in poor efficiency state, such as,
The utilization rate of 4 resource of processor that processing this document uses is only 1% and/or the utilization of memory source that uses of processing this document
Rate not 2%, naturally it is also possible to be that to handle the utilization rates of the Internet resources that this document uses be 1% or handling this document uses
The utilization rates of hard disk resources be 1%, i.e., the one or more in above-mentioned computing resource are less than default 6th threshold value, and
Pre-set 6th threshold value is 5%, at this point it is possible to the total quantity of execution unit 3 is reduced, so that above-mentioned 4 resource of processor,
One or more of utilization rates in memory source, Internet resources or hard disk resources rise to normal model more than 6th threshold value
Enclose, meanwhile, the execution unit 3 for adjusting out can be used for handling other calculating tasks, so as to improve the whole utilization of resource
Rate.
Certainly, during actual use, during file is handled, Internet resources and/or hard disk resources are found
, can also be by adjusting the configuration of at least one execution unit 3, so that Internet resources and/or hard disk resources when being unsatisfactory for requiring
It disclosure satisfy that use demand, and then improve the treatment effeciency of processing this document.
Based on the design identical with above-mentioned computing resource optimization method, as shown in Fig. 2, the embodiment of the present invention additionally provides one
Kind computing resource optimization device, including：
Acquiring unit 1, it is used to obtain the first calculating task handled in calculate node at least one calculating task and accounts for
With the use information of computing resource；
Adjustment unit 2, it is used to be based on the use information, adjusts the calculate node and handles first calculating task
Configuration information.
Preferably, wherein, the computing resource includes at least one of：
4 resource of processor；
Memory source；
Internet resources；
Hard disk resources.
Preferably, described device further includes：
Execution unit 3, first calculating task are handled by least one execution unit 3, and the execution unit 3 is matched somebody with somebody
Processing core and memory source are equipped with, the adjustment calculate node handles the configuration information of first calculating task, bag
Include：
Adjust the configuration of at least one execution unit 3.
Preferably, the adjustment unit 2 is used at least one of：
Adjust the total quantity of the execution unit 3；With
Adjust the processing core quantity of the execution unit 3；With
Adjust the memory source size of the execution unit 3.
Preferably, the adjustment unit 2, it is used to be based on the use information, adjusts the calculate node processing institute
State the configuration information of the first calculating task, including at least one of：
When 4 resource utilization of processor is less than first threshold, the processing core quantity of the execution unit 3 is reduced；With
When 4 resource utilization of processor is higher than second threshold, increase the process cores calculation of the execution unit 3
Amount；With
When memory resource utilization is less than three threshold values, the memory source size of the execution unit 3 is reduced；With
When the memory source utilization rate is higher than four threshold values, increase the memory source size of the execution unit 3；With
When the ratio in high usage state is higher than five threshold values in the computing resource, increase the execution unit
3 total quantity；With
When the ratio in poor efficiency state is higher than six threshold values in the computing resource, the execution unit is reduced
3 total quantity；With
When the Internet resources and/or hard disk resources are unsatisfactory for requiring, matching somebody with somebody at least one execution unit 3 is adjusted
Put.
Equally based on the design identical with above-mentioned computing resource optimization method, as shown in figure 3, present invention also offers one kind
Server cluster, including：
Processor 4；
Memory 5, wherein, executable instruction is stored with the memory 5, wherein, in the executable instruction by institute
When stating the execution of processor 4 so that the processor 4 proceeds as follows：
Obtain the use that the first calculating task handled in calculate node at least one calculating task takes computing resource
Information；
Based on the use information, the configuration information that the calculate node handles first calculating task is adjusted.
Preferably, wherein, the computing resource includes at least one of：
4 resource of processor；
Memory source；
Internet resources；
Hard disk resources.
Preferably, during the processor 4 carries out aforesaid operations, first calculating task is held by least one
Row unit 3 is handled, and the execution unit 3 is configured with processing core and memory source, the adjustment calculate node processing institute
The configuration information of the first calculating task is stated, including：
Adjust the configuration of at least one execution unit 3.
Preferably, the processor 4, which carries out aforesaid operations, includes at least one of：
Adjust the total quantity of the execution unit 3；With
Adjust the processing core quantity of the execution unit 3；With
Adjust the memory source size of the execution unit 3.
Preferably, being based on the use information, the configuration that the calculate node handles first calculating task is adjusted
Information, including at least one of：
When 4 resource utilization of processor is less than first threshold, the processing core quantity of the execution unit 3 is reduced；With
When 4 resource utilization of processor is higher than second threshold, increase the process cores calculation of the execution unit 3
Amount；With
When memory resource utilization is less than three threshold values, the memory source size of the execution unit 3 is reduced；With
When the memory source utilization rate is higher than four threshold values, increase the memory source size of the execution unit 3；With
When the ratio in high usage state is higher than five threshold values in the computing resource, increase the execution unit
3 total quantity；With
When the ratio in poor efficiency state is higher than six threshold values in the computing resource, the execution unit is reduced
3 total quantity；With
When the Internet resources and/or hard disk resources are unsatisfactory for requiring, matching somebody with somebody at least one execution unit 3 is adjusted
Put.
The computing resource optimization device and server cluster introduced by the present embodiment are to be calculated in the embodiment of the present application
Device and server cluster corresponding to method for optimizing resources, so, based on calculating method for optimizing resources in the embodiment of the present application,
Those skilled in the art can understand the specific implementation that resource optimization device and server cluster are calculated in the embodiment of the present application
Mode and its various change form, are just no longer situated between in detail so optimizing device and server cluster for the computing resource herein
Continue.As long as technical staff described in this area implements to calculate the device or server set of method for optimizing resources in the embodiment of the present application
Group, belongs to the scope to be protected of the application.
It should be understood by those skilled in the art that, embodiments herein can be provided as method, system or computer program
Product.Therefore, the application can use the reality in terms of complete hardware embodiment, complete software embodiment or combination software and hardware
Apply the form of example.Moreover, the application can use the computer for wherein including computer usable program code in one or more
The computer program production that usable storage medium is implemented on (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of product.
The application is with reference to the flow according to the method for the embodiment of the present application, equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that it can be realized by computer program instructions every first-class in flowchart and/or the block diagram
The combination of flow and/or square frame in journey and/or square frame and flowchart and/or the block diagram.These computer programs can be provided
The processor of all-purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices is instructed to produce
A raw machine so that the instruction performed by computer or the processor of other programmable data processing devices, which produces, to be used in fact
The dress for the function of being specified in the flow or multiple flows and/or a square frame of block diagram or multiple square frames of present flow chart
Put.
These computer program instructions, which may also be stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that the instruction being stored in the computer-readable memory, which produces, to be included referring to
The manufacture of device is made, which realizes in a flow of flow chart or multiple flows and/or a side of block diagram
The function of being specified in frame or multiple square frames.
These computer program instructions can be also loaded into computer or other programmable data processing devices so that counted
Series of operation steps is performed on calculation machine or other programmable devices to produce computer implemented processing, thus in computer or
The instruction performed on other programmable devices provides the flow or multiple flows and/or block diagram that are used for realization in flow chart
A square frame or multiple square frames in specify function the step of.
Above example is only the exemplary embodiment of the present invention, is not used in the limitation present invention, protection scope of the present invention
It is defined by the claims.Those skilled in the art can make the present invention respectively in the essence and protection domain of the present invention
Kind modification or equivalent substitution, this modification or equivalent substitution also should be regarded as being within the scope of the present invention.
Claims (10)
1. a kind of computing resource optimization method, including：
Obtain the use information that the first calculating task handled in calculate node at least one calculating task takes computing resource；
Based on the use information, the configuration information that the calculate node handles first calculating task is adjusted.
2. computing resource optimization method according to claim 1, wherein, the computing resource includes at least one of：
Processor resource；
Memory source；
Internet resources；
Hard disk resources.
3. computing resource optimization method according to claim 2, wherein, first calculating task is by least one execution
Cell processing, the execution unit are configured with processing core and memory source, the adjustment calculate node processing described the
The configuration information of one calculating task, including：
Adjust the configuration of at least one execution unit.
4. computing resource optimization method according to claim 3, the described method includes at least one of：
Adjust the total quantity of the execution unit；With
Adjust the processing core quantity of the execution unit；With
Adjust the memory source size of the execution unit.
5. computing resource optimization method according to claim 4, described to be based on the use information, adjust described calculate and save
Point handles the configuration information of first calculating task, including at least one of：
When processor resource utilization rate is less than first threshold, the processing core quantity of the execution unit is reduced；With
When the processor resource utilization rate is higher than second threshold, increase the processing core quantity of the execution unit；With
When memory resource utilization is less than three threshold values, the memory source size of the execution unit is reduced；With
When the memory source utilization rate is higher than four threshold values, increase the memory source size of the execution unit；With
When the ratio in high usage state is higher than five threshold values in the computing resource, increase the total of the execution unit
Quantity；With
When the ratio in poor efficiency state is higher than six threshold values in the computing resource, the total of the execution unit is reduced
Quantity；With
When the Internet resources and/or hard disk resources are unsatisfactory for requiring, the configuration of at least one execution unit is adjusted.
6. a kind of computing resource optimizes device, including：
Acquiring unit, it is used to obtain the first calculating task handled in calculate node at least one calculating task and takes calculating
The use information of resource；
Adjustment unit, it is used to be based on the use information, adjusts the calculate node and handles matching somebody with somebody for first calculating task
Confidence ceases.
7. computing resource according to claim 6 optimizes device, wherein, the computing resource includes at least one of：
Processor resource；
Memory source；
Internet resources；
Hard disk resources.
8. computing resource according to claim 7 optimizes device, further include：
Execution unit, first calculating task are handled by least one execution unit, and the execution unit is configured with place
Core and memory source are managed, the adjustment calculate node handles the configuration information of first calculating task, including：
Adjust the configuration of at least one execution unit.
9. computing resource according to claim 8 optimizes device, the adjustment unit is used at least one of：
Adjust the total quantity of the execution unit；With
Adjust the processing core quantity of the execution unit；With
Adjust the memory source size of the execution unit.
10. a kind of server cluster, including：
Processor；
Memory, wherein, executable instruction is stored with the memory, wherein, in the executable instruction by the processing
When device performs so that the processor proceeds as follows：
Obtain the use information that the first calculating task handled in calculate node at least one calculating task takes computing resource；
Based on the use information, the configuration information that the calculate node handles first calculating task is adjusted.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711405545.0A CN108021453A (en) | 2017-12-22 | 2017-12-22 | A kind of computing resource optimization method, device and server cluster |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711405545.0A CN108021453A (en) | 2017-12-22 | 2017-12-22 | A kind of computing resource optimization method, device and server cluster |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108021453A true CN108021453A (en) | 2018-05-11 |
Family
ID=62074404
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711405545.0A Pending CN108021453A (en) | 2017-12-22 | 2017-12-22 | A kind of computing resource optimization method, device and server cluster |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108021453A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105718364A (en) * | 2016-01-15 | 2016-06-29 | 西安交通大学 | Dynamic assessment method for ability of computation resource in cloud computing platform |
| CN106897132A (en) * | 2017-02-27 | 2017-06-27 | 郑州云海信息技术有限公司 | The method and device of a kind of server task scheduling |
| WO2017133351A1 (en) * | 2016-02-05 | 2017-08-10 | 华为技术有限公司 | Resource allocation method and resource manager |
-
2017
- 2017-12-22 CN CN201711405545.0A patent/CN108021453A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105718364A (en) * | 2016-01-15 | 2016-06-29 | 西安交通大学 | Dynamic assessment method for ability of computation resource in cloud computing platform |
| WO2017133351A1 (en) * | 2016-02-05 | 2017-08-10 | 华为技术有限公司 | Resource allocation method and resource manager |
| CN107045456A (en) * | 2016-02-05 | 2017-08-15 | 华为技术有限公司 | A kind of resource allocation methods and explorer |
| CN106897132A (en) * | 2017-02-27 | 2017-06-27 | 郑州云海信息技术有限公司 | The method and device of a kind of server task scheduling |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109002358B (en) | Mobile terminal software self-adaptive optimization scheduling method based on deep reinforcement learning | |
| CN105993024B (en) | For the classical simulation constant of quantum chemical modelling and sequence | |
| CN104239141A (en) | Task optimized-scheduling method in data center on basis of critical paths of workflow | |
| Chen et al. | Learning-based resource allocation in cloud data center using advantage actor-critic | |
| CN107332902A (en) | User's request distribution method, device and the computing device of online customer service system | |
| CN107621973A (en) | A kind of method for scheduling task and device across cluster | |
| CN104536832B (en) | A kind of virtual machine deployment method | |
| WO2015144008A1 (en) | Method and device for allocating physical machine to virtual machine | |
| CN108055292B (en) | Optimization method for mapping from virtual machine to physical machine | |
| Wang et al. | Job scheduling for large-scale machine learning clusters | |
| CN109525500A (en) | A kind of information processing method and information processing unit of self-adjusting threshold value | |
| CN106202092A (en) | The method and system that data process | |
| CN109976901A (en) | A kind of resource regulating method, device, server and readable storage medium storing program for executing | |
| CN110347515A (en) | A kind of resource optimal distribution method of suitable edge calculations environment | |
| CN107608788A (en) | A kind of control method, device and equipment | |
| CN107872405A (en) | Distributed bandwidth allocation and regulation | |
| CN103106112A (en) | Method and device based on maximum load and used for load balancing scheduling | |
| CN111601328B (en) | Mobile edge computing device and method | |
| CN106855825A (en) | A kind of task processing method and its device | |
| Amalarethinam et al. | Workflow scheduling for public cloud using genetic algorithm (WSGA) | |
| CN108021453A (en) | A kind of computing resource optimization method, device and server cluster | |
| CN110347477B (en) | Service self-adaptive deployment method and device in cloud environment | |
| CN105988952B (en) | The method and apparatus for distributing hardware-accelerated instruction for Memory Controller Hub | |
| KR101639003B1 (en) | Manicore system based cpu/gpu and method for distributing workload for cpu/gpu concurrent processing | |
| US20200314019A1 (en) | Managing bandwidth based on user behavior |
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: 20180511 |