CN108021447A - A kind of method and system that optimal resource policy is determined based on distributed data - Google Patents
A kind of method and system that optimal resource policy is determined based on distributed data Download PDFInfo
- Publication number
- CN108021447A CN108021447A CN201711217710.XA CN201711217710A CN108021447A CN 108021447 A CN108021447 A CN 108021447A CN 201711217710 A CN201711217710 A CN 201711217710A CN 108021447 A CN108021447 A CN 108021447A
- Authority
- CN
- China
- Prior art keywords
- resource
- microserver
- current
- rise
- intelligent training
- 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.)
- Granted
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
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/505—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the load
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention relates to a kind of method and system that optimal resource policy is determined based on distributed data, this method includes:Current resource information is gathered, and establishes intelligent training model;Then the use resource rate of rise of each microserver in current any time period is calculated, and intelligent training model is optimized;Then calculating each microserver in following any time period needs stock number to be used;Repeat the above process until calculating current equal with last use resource rate of rise using resource rate of rise;Secondly the Node distribution strategy protocol of microserver optimized is determined;Control instruction is ultimately produced, node resource is adjusted in real time.Further relate to a kind of system.Resource information can be obtained according to real-time by the present invention, and automated according to resource information, intelligently calculate optimal distribution strategy, distribute micro services according to optimal strategy needed for resource, allow resource to save resources costs using maximizing for enterprise.
Description
Technical field
The invention belongs to microserver field, more particularly to a kind of side that optimal resource policy is determined based on distributed data
Method and system.
Background technology
Currently existing micro services system has relative independentability, does not form a whole set of service network, not with tune
With frequency, monitoring service state, the bulk information for calculating flow of services comes so as to carry out intelligence, draws real-time optimal resource
Scheduling is using strategy, and the resource needed for can not distributing micro services according to optimal strategy, resource is allowed using maximizing, for enterprise
Industry saves resources costs.
The content of the invention
The technical problems to be solved by the invention are:Existing micro services system can not show that real-time optimal scheduling of resource makes
With strategy, and the resource needed for microserver can not be distributed according to optimal strategy, allow resource to be saved using maximizing for enterprise
Resource-saving cost.
To solve technical problem above, optimal resource policy is determined based on distributed data the present invention provides a kind of
Method, this method include:
S1, gathers the Current resource information of at least two microserver, and establishes intelligent training model;
S2, according to the Current resource information and the intelligent training model, calculates each micro- in current any time period
The use resource rate of rise of server, and resource rate of rise is used according to described, the intelligent training model is carried out excellent
Change;
S3, according to the intelligent training model using after resource rate of rise and optimization, calculates and will appoint in future
Each microserver needs stock number to be used in one period;
S4, repeats step S1-S3, until calculating current described using resource rate of rise and last institute
State and use resource rate of rise equal, then perform step S5;
S5, needs stock number to be used, really according to the intelligent training model after optimization and each microserver
Surely the Node distribution strategy protocol of the microserver optimized;
S6, according to the Node distribution strategy protocol, generates control instruction, at the same according to the control instruction to it is described extremely
The node resource of few 2 microservers is adjusted in real time.
Beneficial effects of the present invention:By above-mentioned method, resource information can be obtained according to real-time, and believe according to resource
Breath automation, intelligently calculate optimal distribution strategy, distribute microserver according to optimal strategy needed for resource, allow
Resource saves resources costs using maximizing, for enterprise.
Further, the Current resource information includes:The ip resources of each microserver, cpu memories, load pressure and
Network traffic information, the stock number taken and actual negative carrying capacity sf.
Further, preset in intelligent training model at least two microserver total resources and it is each in incognito
The load capacity yf of business device.
Further, the S3 includes:When the ratio of the actual negative carrying capacity sf and load capacity yf is more than 0.8
When, then according to the intelligent training model using after resource rate of rise and optimization, calculate in following any time
Each microserver needs stock number to be used in section.
Further, further included in the S3:When the ratio of the actual negative carrying capacity sf and load capacity yf is less than 0.2
When, then according to the intelligent training model using after resource rate of rise and optimization, calculate in following any time period
The Resource recovery of interior each microserver or the stock number for discharging resource, wherein the Resource recovery refers to itself carry currently
When the stock number of confession is insufficient, the resource of other microserver releases is obtained;The release resource refers to itself provide current
Stock number it is excessive when, release resource gives other microservers.
Above-mentioned further beneficial effect:According to using the intelligent training model after resource rate of rise and optimization, determine
The Resource recovery of each microserver or the stock number for discharging resource, are so conducive to subsequently according to Node distribution strategy protocol
Resource is allocated.
The invention further relates to a kind of system that optimal resource policy is determined based on distributed data, which includes:
Mongodb big datas cluster, big data calculate center, intelligence computation center, policy calculation center, micro services cluster;
The micro services cluster is to include:At least two microserver, wherein at least two microserver forms distribution
Formula structure;
The mongodb big datas cluster, for using at least two microserver in the micro services cluster
Current resource information;
The big data calculates center, for the Current resource in quick real-time query mongodb big data clusters
Information, the intelligence computation center is submitted to by the Current resource information;
The intelligence computation center, for establishing intelligent training model, and according to the Current resource information and the intelligence
Energy training pattern, calculates the use resource rate of rise of each microserver in current any time period, and according to the use
Resource rate of rise, optimizes the intelligent training model;It is additionally operable to use resource rate of rise and optimization according to described
The intelligent training model afterwards, calculating each microserver in following any time period needs stock number to be used;And
Stock number to be used is needed according to the intelligent training model after optimization and each microserver, determines to optimize micro-
The Node distribution strategy protocol of server;
The policy calculation center, for according to the Node distribution strategy protocol, generating control instruction, while according to institute
Control instruction is stated to adjust the node resource of at least two microserver in real time.
Beneficial effects of the present invention:By above-mentioned system, resource information can be obtained according to real-time, and believe according to resource
Breath automation, intelligently calculate optimal distribution strategy, distribute microserver according to optimal strategy needed for resource, allow
Resource saves resources costs using maximizing, for enterprise.
Further, the Current resource information includes:The ip resources of each microserver, cpu memories, load pressure and
Network traffic information, the stock number taken and actual negative carrying capacity sf.
Further, preset in intelligent training model at least two microserver total resources and it is each in incognito
The load capacity yf of business device.
Further, the intelligence computation center, is specifically additionally operable to as the actual negative carrying capacity sf and load capacity yf
Ratio be more than 0.8 when, then according to it is described using resource rate of rise and optimization after the intelligent training model, calculate
Each microserver needs stock number to be used in following any time period.
Further, the intelligence computation center, is specifically additionally operable to as the actual negative carrying capacity sf and load capacity yf
Ratio be less than 0.2 when, then according to it is described using resource rate of rise and optimization after the intelligent training model, calculate not
Carry out the recycling of each microserver in any time period or discharge the stock number of resource, wherein the Resource recovery refers to working as
When the stock number of itself preceding offer is insufficient, the resource of other microservers releases is obtained;The release resource refers to current
When the stock number itself provided is excessive, release resource gives other microservers.
Above-mentioned further beneficial effect:According to using the intelligent training model after resource rate of rise and optimization, determine
The Resource recovery of each microserver or the stock number for discharging resource, are so conducive to subsequently according to Node distribution strategy protocol
Resource is allocated.
Brief description of the drawings
Fig. 1 is a kind of flow chart of method that optimal resource policy is determined based on distributed data of the present invention;
Fig. 2 is a kind of schematic diagram of system that optimal resource policy is determined based on distributed data of the present invention.
Embodiment
The principle and features of the present invention will be described below with reference to the accompanying drawings, and the given examples are served only to explain the present invention, and
It is non-to be used to limit the scope of the present invention.
A kind of optimal resource policy is determined based on distributed data as shown in Figure 1, being provided in the embodiment of the present invention 1
Method, this method include:
S1, gathers the Current resource information of at least two microserver, and establishes intelligent training model;
S2, according to the Current resource information and the intelligent training model, calculates each micro- in current any time period
The use resource rate of rise of server, and resource rate of rise is used according to described, the intelligent training model is carried out excellent
Change;
S3, according to the intelligent training model using after resource rate of rise and optimization, calculates and will appoint in future
Each microserver needs stock number to be used in one period;
S4, repeats step S1-S3, until calculating current described using resource rate of rise and last institute
State and use resource rate of rise equal, then perform step S5;
S5, needs stock number to be used, really according to the intelligent training model after optimization and each microserver
Surely the Node distribution strategy protocol of the microserver optimized;
S6, according to the Node distribution strategy protocol, generates control instruction, at the same according to the control instruction to it is described extremely
The node resource of few 2 microservers is adjusted in real time.
It should be noted that it is that some when currently running for gathering more microservers provide first in the present embodiment 1
Source information, these resource informations are all the capability operations for representing that microserver is current, such as:The hardware resource letter of microserver
Breath, the network information, application service information, application tracking information etc., while also establish intelligent training model;Establish intelligent training
After model, intelligent training model can calculate current be not fixed in the period often according to some resource informations of microserver
The use resource rate of rise of a microserver, in addition can according to these using resource rate of rise to intelligent training model into
Row optimization (i.e. according to the parameter using resource rate of rise adjustment intelligent training model), so using the intelligent training after optimization
Model and go to calculate using resource rate of rise and each microserver in the period will be not fixed in future needs resource to be used
Amount;By constantly obtain microserver renewal after resource information, continue to be adjusted optimization to intelligent training model, until
It is current equal with last use resource rate of rise using resource rate of rise when calculating, show current micro services
Device is in stable state, and intelligent training model falls within the state more optimized at present, therefore, can be according to optimization after
Intelligent training model and each microserver need stock number to be used, determine the Node distribution plan of microserver optimized
Slightly scheme, thus can be according to Node distribution strategy protocol, and adjusting each microserver in time needs stock number to be used,
Resource is fully distributed and is fully used by each microserver, allow resource to be saved using maximization is realized for enterprise
Resource-saving cost.
Alternatively, Current resource information includes described in another embodiment 2:In the ip resources of each microserver, cpu
Deposit, load pressure and network traffic information, the stock number taken and actual negative carrying capacity sf.
It should be noted that the present embodiment 2 is the refinement explanation carried out on the basis of above-described embodiment 1, in this reality
Apply in example 2 want parsing be to be mentioned to Current resource information in embodiment 1 all to include:The ip resources of each microserver,
Cpu memories, load pressure and network traffic information, the stock number taken and actual negative carrying capacity sf, these information are all to fill
Divide the operating condition of the current microserver of description, be conducive to subsequently transfer these microservers.
Alternatively, the total resources of at least two microserver are preset in intelligent training model in another embodiment 3
The load capacity yf of amount and each microserver.
It should be noted that the present embodiment 3 is the explanation and improvement carried out on the basis of above-described embodiment 2, this implementation
Example 3 is to demonstrate the need for presetting the total resources of all microservers in intelligent training model in advance and preset each microserver
Load capacity yf, when being so to subsequently be calculated, the stock number and reality that are taken with each microserver for actually measuring
Border load capacity sf is compared, and judges the situation of microserver.
Alternatively, S3 includes described in another embodiment 4:When the actual negative carrying capacity sf's and load capacity yf
When ratio is more than 0.8, then according to the intelligent training model using after resource rate of rise and optimization, calculate not
Carrying out each microserver in any time period needs stock number to be used.
It should be noted that the technical solution in the present embodiment 4 is improved in the technology of above-described embodiment 3, this
It is that the actual negative carrying capacity sf in embodiment 3 and load capacity yf is subjected to ratio in judgement in embodiment 4, as actual negative carrying capacity sf and institute
When stating the ratio of load capacity yf more than 0.8, the use for illustrating the current microserver is that comparison is sufficient, then may determine that it
In follow-up any time period the current microserver probably also can behaviour in service can also continue for some time, may infer that with this
Going out microserver in following any time period needs stock number to be used.
Alternatively, further included described in another embodiment 5 in S3:As the actual negative carrying capacity sf and the load capacity yf
Ratio be less than 0.2 when, then according to it is described using resource rate of rise and optimization after the intelligent training model, calculate not
Carry out the stock number of the Resource recovery of each microserver or release resource in any time period, wherein the Resource recovery refers to
When the current stock number itself provided is insufficient, the resource of other microserver releases is obtained;The release resource refers to
When the stock number of itself current offer is excessive, release resource gives other microservers.
It should be noted that the present embodiment 5 is the scheme carried out in the technology of above-described embodiment 3, it is in the present embodiment 4
Actual negative carrying capacity sf in embodiment 3 and load capacity yf is subjected to ratio in judgement, as actual negative carrying capacity sf and the load capacity yf
Ratio when being less than 0.2, the use for illustrate the current microserver is that utilization rate is bad, then when may determine that it is subsequently any
Between in section the current microserver need to be improved the server, for example improve the utilization rate of the microserver, fully make
With resource, the recycling of calculating each microserver in following any time period in need or the stock number of release resource,
Wherein Resource recovery refers to the resource for when the current stock number itself provided is insufficient, obtaining other microserver releases;Release
Put resource and refer to that release resource gives other microservers when the current stock number itself provided is excessive.In the present embodiment 5
According to using the intelligent training model after resource rate of rise and optimization, determine the recycling of each microserver or discharge resource
Stock number, be so conducive to subsequently to allocate resource according to Node distribution strategy protocol that (such as resource reclaim is to instruction
(stop micro services application, delete the instruction such as application cache data, the node of micro services container is adjusted in real time, to reach
The optimization use of resource).
As shown in Fig. 2, further related in the embodiment of the present invention 6 a kind of be based on what distributed data determined optimal resource policy
System, the system include:Mongodb big datas cluster, big data calculate center, intelligence computation center, policy calculation center, in incognito
Business cluster;
The micro services cluster is to include:At least two microserver, wherein at least two microserver forms distribution
Formula structure;
The mongodb big datas cluster, for working as using at least two micro services in the micro services cluster
Preceding resource information;
The big data calculates center, is calculated for quick real-time query described current in mongodb big data clusters
Resource information, the intelligence computation center is submitted to by the Current resource information;
The intelligence computation center, for establishing intelligent training model, and according to the Current resource information and the intelligence
Energy training pattern, calculates the use resource rate of rise of each microserver in current any time period, and according to the use
Resource rate of rise, optimizes the intelligent training model;It is additionally operable to use resource rate of rise and optimization according to described
The intelligent training model afterwards, calculating each microserver in following any time period needs stock number to be used;And
Stock number to be used is needed according to the intelligent training model after optimization and each microserver, determines to optimize micro-
The Node distribution strategy protocol of server;
The policy calculation center, for according to the Node distribution strategy protocol, generating control instruction, while according to institute
Control instruction is stated to adjust the node resource of at least two microserver in real time.
It should be noted that it is that some when currently running for gathering more microservers provide first in the present embodiment 6
Source information, these resource informations are all the capability operations for representing that microserver is current, such as:The hardware resource letter of microserver
Breath, the network information, application service information, application tracking information etc., while also establish intelligent training model;Establish intelligent training
After model, intelligent training model can calculate current be not fixed in the period often according to some resource informations of microserver
The use resource rate of rise of a microserver, in addition can according to these using resource rate of rise to intelligent training model into
Row optimization (i.e. according to the parameter using resource rate of rise adjustment intelligent training model), so using the intelligent training after optimization
Model and go to calculate using resource rate of rise and each microserver in the period will be not fixed in future needs resource to be used
Amount;By constantly obtain microserver renewal after resource information, continue to be adjusted optimization to intelligent training model, until
It is current equal with last use resource rate of rise using resource rate of rise when calculating, show current micro services
Device is in stable state, and intelligent training model falls within the state more optimized at present, therefore, can be according to optimization after
Intelligent training model and each microserver need stock number to be used, determine the Node distribution plan of microserver optimized
Slightly scheme, thus can be according to Node distribution strategy protocol, and adjusting each microserver in time needs stock number to be used,
Resource is fully distributed and is fully used by each microserver, allow resource to be saved using maximization is realized for enterprise
Resource-saving cost
Alternatively, Current resource information includes described in another embodiment 7:In the ip resources of each microserver, cpu
Deposit, load pressure and network traffic information, the stock number taken and actual negative carrying capacity sf.
It should be noted that the present embodiment 7 is the refinement explanation carried out on the basis of above-described embodiment 6, in this reality
Apply in example 7 want parsing be to be mentioned to Current resource information in embodiment 6 all to include:The ip resources of each microserver,
Cpu memories, load pressure and network traffic information, the stock number taken and actual negative carrying capacity sf, these information are all to fill
Divide the operating condition of the current microserver of description, be conducive to subsequently carry out transfer to these microservers
Alternatively, the total resources of at least two microserver are preset in intelligent training model in another embodiment 8
The load capacity yf of amount and each microserver.
It should be noted that the present embodiment 8 is the explanation and improvement carried out on the basis of above-described embodiment 7, this implementation
Example 8 is to demonstrate the need for presetting the total resources of all microservers in intelligent training model in advance and preset each microserver
Load capacity yf, when being so to subsequently be calculated, the stock number and reality that are taken with each microserver for actually measuring
Border load capacity sf is compared, and judges the situation of microserver.
Alternatively, the intelligence computation center described in another embodiment 9, be specifically additionally operable to when the actual negative carrying capacity sf with
When the ratio of the load capacity yf is more than 0.8, then according to the intelligent training using after resource rate of rise and optimization
Model, calculating each microserver in following any time period needs stock number to be used.
It should be noted that the technical solution in the present embodiment 9 is improved in the technology of above-described embodiment 8, this
It is that the actual negative carrying capacity sf in embodiment 8 and load capacity yf is subjected to ratio in judgement in embodiment 9, as actual negative carrying capacity sf and institute
When stating the ratio of load capacity yf more than 0.8, the use for illustrating the current microserver is that comparison is sufficient, then may determine that it
In follow-up any time period the current microserver probably also can behaviour in service can also continue for some time, may infer that with this
Going out microserver in following any time period needs stock number to be used.
Alternatively, the intelligence computation center described in another embodiment 10, is specifically additionally operable to work as the actual negative carrying capacity sf
When being less than 0.2 with the ratio of the load capacity yf, then instructed according to the intelligence using after resource rate of rise and optimization
Practice model, calculate the recycling of each microserver in following any time period or discharge the stock number of resource, wherein described
Resource recovery refers to the resource for when the current stock number itself provided is insufficient, obtaining other microserver releases;It is described to release
Put resource and refer to that release resource gives other microservers when the current stock number itself provided is excessive.
It should be noted that the present embodiment 10 is the scheme carried out in the technology of above-described embodiment 8, in the present embodiment 10
It is that the actual negative carrying capacity sf in embodiment 8 and load capacity yf is subjected to ratio in judgement, as actual negative carrying capacity sf and the load capacity
When the ratio of yf is less than 0.2, the use for illustrating the current microserver is that utilization rate is bad, then may determine that it is follow-up any
The current microserver needs to be improved the server in period, for example improves the utilization rate of the microserver, fully
Using resource, the recycling of calculating each microserver in following any time period in need or the resource of release resource
Amount, wherein Resource recovery refer to the resource for when the current stock number itself provided is insufficient, obtaining other microserver releases;
Release resource refers to that release resource gives other microservers when the current stock number itself provided is excessive.In the present embodiment
According to using the intelligent training model after resource rate of rise and optimization in 10, the recycling or release of each microserver are determined
The stock number of resource, be so conducive to subsequently to allocate resource according to Node distribution strategy protocol (such as resource reclaim arrives
Instruction (stop micro services application, delete the instruction such as application cache data, the node of micro services container is adjusted in real time, with
Reach the optimization use of resource).
In the present specification, a schematic expression of the above terms does not necessarily refer to the same embodiment or example.
Moreover, particular features, structures, materials, or characteristics described can be in any one or more of the embodiments or examples with suitable
Mode combines.In addition, without conflicting with each other, those skilled in the art can be by the difference described in this specification
Embodiment or example and different embodiments or exemplary feature are combined and combine.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent replacement, improvement and so on, should all be included in the protection scope of the present invention.
Claims (10)
- A kind of 1. method that optimal resource policy is determined based on distributed data, it is characterised in that this method includes:S1, gathers the Current resource information of at least two microserver, and establishes intelligent training model;S2, according to the Current resource information and the intelligent training model, calculates each micro services in current any time period The use resource rate of rise of device, and resource rate of rise is used according to described, the intelligent training model is optimized;S3, according to the intelligent training model using after resource rate of rise and optimization, calculates when future is any Between in section each microserver need stock number to be used;S4, repeats step S1-S3, until calculate it is current it is described using resource rate of rise with it is last described in make It is equal with resource rate of rise, then perform step S5;S5, needs stock number to be used according to the intelligent training model after optimization and each microserver, determines most The Node distribution strategy protocol of the microserver of optimization;S6, according to the Node distribution strategy protocol, generates control instruction, while according to the control instruction to described at least 2 The node resource of a microserver is adjusted in real time.
- 2. according to the method described in claim 1, it is characterized in that, the Current resource information includes:The i of each microserver P resources, cpu memories, load pressure and network traffic information, the stock number taken and actual negative carrying capacity sf.
- 3. according to the method described in claim 2, it is characterized in that, preset described at least two in incognito in intelligent training model Be engaged in the total resources of device and the load capacity yf of each microserver.
- 4. according to the method described in claim 3, it is characterized in that, the S3 includes:When the actual negative carrying capacity sf and institute When stating the ratio of load capacity yf more than 0.8, then resource rate of rise and the intelligent training mould after optimization are used according to described Type, calculating each microserver in following any time period needs stock number to be used.
- 5. according to the method described in claim 3, it is characterized in that, further included in the S3:When the actual negative carrying capacity sf with When the ratio of the load capacity yf is less than 0.2, then according to the intelligent training using after resource rate of rise and optimization Model, calculates the stock number of the Resource recovery of each microserver or release resource in following any time period, wherein institute State the resource that Resource recovery refers to when the current stock number itself provided is insufficient, obtain other microserver releases;It is described Release resource refers to that release resource gives other microservers when the current stock number itself provided is excessive.
- 6. a kind of system that optimal resource policy is determined based on distributed data, it is characterised in that the system includes:mongodb Big data cluster, big data calculate center, intelligence computation center, policy calculation center, micro services cluster;The micro services cluster is to include:At least two microserver, wherein at least two microserver forms distributed knot Structure;The mongodb big datas cluster, for using the current of at least two microserver in the micro services cluster Resource information;The big data calculates center, for the Current resource information in quick real-time query mongodb big data clusters, The Current resource information is submitted into the intelligence computation center;The intelligence computation center, is instructed for establishing intelligent training model, and according to the Current resource information and the intelligence Practice model, calculate the use resource rate of rise of each microserver in current any time period, and resource is used according to described Rate of rise, optimizes the intelligent training model;Be additionally operable to according to it is described using resource rate of rise and optimization after The intelligent training model, calculating each microserver in following any time period needs stock number to be used;And according to The intelligent training model and each microserver after optimization need stock number to be used, determine the micro services optimized The Node distribution strategy protocol of device;The policy calculation center, for according to the Node distribution strategy protocol, generating control instruction, while according to the control System instruction adjusts the node resource of at least two microserver in real time.
- 7. system according to claim 6, it is characterised in that the Current resource information includes:The i of each microserver P resources, cpu memories, load pressure and network traffic information, the stock number taken and actual negative carrying capacity sf.
- 8. system according to claim 7, it is characterised in that preset described at least two in incognito in intelligent training model Be engaged in the total resources of device and the load capacity yf of each microserver.
- 9. system according to claim 8, it is characterised in that the intelligence computation center, is specifically additionally operable to work as the reality When the ratio of border load capacity sf and the load capacity yf are more than 0.8, then according to it is described using resource rate of rise and optimization after The intelligent training model, calculating each microserver in following any time period needs stock number to be used.
- 10. system according to claim 8, it is characterised in that the intelligence computation center, is specifically additionally operable to work as the reality When the ratio of border load capacity sf and the load capacity yf are less than 0.2, then according to it is described using resource rate of rise and optimization after The intelligent training model, calculates the money of the Resource recovery of each microserver or release resource in following any time period Source is measured, wherein the Resource recovery refers to, when the current stock number itself provided is insufficient, obtain other microserver releases Resource;The release resource refers to that release resource gives other microservers when the current stock number itself provided is excessive.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711217710.XA CN108021447B (en) | 2017-11-28 | 2017-11-28 | Method and system for determining optimal resource strategy based on distributed data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711217710.XA CN108021447B (en) | 2017-11-28 | 2017-11-28 | Method and system for determining optimal resource strategy based on distributed data |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108021447A true CN108021447A (en) | 2018-05-11 |
CN108021447B CN108021447B (en) | 2021-08-03 |
Family
ID=62077243
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711217710.XA Active CN108021447B (en) | 2017-11-28 | 2017-11-28 | Method and system for determining optimal resource strategy based on distributed data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108021447B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109005223A (en) * | 2018-07-26 | 2018-12-14 | 南京邮电大学 | Internet of Things resource regulating method and system, computer readable storage medium and terminal |
CN109150826A (en) * | 2018-06-29 | 2019-01-04 | 中译语通科技股份有限公司 | A method of optimization presence information load |
CN109639598A (en) * | 2018-10-19 | 2019-04-16 | 深圳平安财富宝投资咨询有限公司 | Request processing method, server, storage medium and device based on micro services |
CN110971623A (en) * | 2018-09-28 | 2020-04-07 | 中兴通讯股份有限公司 | Micro-service instance elastic scaling method and device and storage medium |
WO2020075017A1 (en) * | 2018-10-12 | 2020-04-16 | International Business Machines Corporation | Auto tuner for cloud micro services embeddings |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030073064A1 (en) * | 2001-10-12 | 2003-04-17 | Lee Riggs | Methods and systems for registering and authenticating recipients of training provided through data communications networks to remote electronic devices |
US20060229896A1 (en) * | 2005-04-11 | 2006-10-12 | Howard Rosen | Match-based employment system and method |
CN105187512A (en) * | 2015-08-13 | 2015-12-23 | 航天恒星科技有限公司 | Method and system for load balancing of virtual machine clusters |
CN205692107U (en) * | 2016-05-13 | 2016-11-16 | 威海明达创新科技有限公司 | A kind of cluster micro server system of distributed structure/architecture |
CN106557370A (en) * | 2016-11-28 | 2017-04-05 | 上海宝尊电子商务有限公司 | Computing resource dynamic dispatching platform based on Realtime Statistics |
CN106648890A (en) * | 2016-12-06 | 2017-05-10 | 中国科学院重庆绿色智能技术研究院 | Cloud computing server resource on-line management method and system with energy consumption sensing function |
CN107042802A (en) * | 2016-12-13 | 2017-08-15 | 象翌微链科技发展有限公司 | A kind of system and method for control engine condition |
CN107124630A (en) * | 2017-03-30 | 2017-09-01 | 华为技术有限公司 | The method and device of node data management |
CN107343164A (en) * | 2016-04-29 | 2017-11-10 | 中国电信股份有限公司 | The distribution method and device of vision control task |
-
2017
- 2017-11-28 CN CN201711217710.XA patent/CN108021447B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030073064A1 (en) * | 2001-10-12 | 2003-04-17 | Lee Riggs | Methods and systems for registering and authenticating recipients of training provided through data communications networks to remote electronic devices |
US20060229896A1 (en) * | 2005-04-11 | 2006-10-12 | Howard Rosen | Match-based employment system and method |
CN105187512A (en) * | 2015-08-13 | 2015-12-23 | 航天恒星科技有限公司 | Method and system for load balancing of virtual machine clusters |
CN107343164A (en) * | 2016-04-29 | 2017-11-10 | 中国电信股份有限公司 | The distribution method and device of vision control task |
CN205692107U (en) * | 2016-05-13 | 2016-11-16 | 威海明达创新科技有限公司 | A kind of cluster micro server system of distributed structure/architecture |
CN106557370A (en) * | 2016-11-28 | 2017-04-05 | 上海宝尊电子商务有限公司 | Computing resource dynamic dispatching platform based on Realtime Statistics |
CN106648890A (en) * | 2016-12-06 | 2017-05-10 | 中国科学院重庆绿色智能技术研究院 | Cloud computing server resource on-line management method and system with energy consumption sensing function |
CN107042802A (en) * | 2016-12-13 | 2017-08-15 | 象翌微链科技发展有限公司 | A kind of system and method for control engine condition |
CN107124630A (en) * | 2017-03-30 | 2017-09-01 | 华为技术有限公司 | The method and device of node data management |
Non-Patent Citations (3)
Title |
---|
DEEPAK C. SHADRACH 等: "A Weighted Metric Based Adaptive Algorithm for Web Server Load Balancing", 《2009 THIRD INTERNATIONAL SYMPOSIUM ON INTELLIGENT INFORMATION TECHNOLOGY APPLICATION》 * |
HONG ZHONG 等: "LBBSRT: An efficient SDN load balancing scheme based on server response time", 《FUTURE GENERATION COMPUTER SYSTEMS》 * |
钟 麟 等: "一种基于 MongoDB 的大数据管理架构", 《计算机(七)网络》 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109150826A (en) * | 2018-06-29 | 2019-01-04 | 中译语通科技股份有限公司 | A method of optimization presence information load |
CN109005223A (en) * | 2018-07-26 | 2018-12-14 | 南京邮电大学 | Internet of Things resource regulating method and system, computer readable storage medium and terminal |
CN110971623A (en) * | 2018-09-28 | 2020-04-07 | 中兴通讯股份有限公司 | Micro-service instance elastic scaling method and device and storage medium |
CN110971623B (en) * | 2018-09-28 | 2023-02-17 | 中兴通讯股份有限公司 | Micro-service instance elastic scaling method and device and storage medium |
WO2020075017A1 (en) * | 2018-10-12 | 2020-04-16 | International Business Machines Corporation | Auto tuner for cloud micro services embeddings |
US10673708B2 (en) | 2018-10-12 | 2020-06-02 | International Business Machines Corporation | Auto tuner for cloud micro services embeddings |
GB2587765A (en) * | 2018-10-12 | 2021-04-07 | Ibm | Auto tuner for cloud micro services embeddings |
GB2587765B (en) * | 2018-10-12 | 2021-08-25 | Ibm | Auto tuner for cloud micro services embeddings |
CN109639598A (en) * | 2018-10-19 | 2019-04-16 | 深圳平安财富宝投资咨询有限公司 | Request processing method, server, storage medium and device based on micro services |
Also Published As
Publication number | Publication date |
---|---|
CN108021447B (en) | 2021-08-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108021447A (en) | A kind of method and system that optimal resource policy is determined based on distributed data | |
US10848428B2 (en) | Method for dynamically allocating resources in an SDN/NFV network based on load balancing | |
LU500343B1 (en) | Self-Adaptive Energy-Optimal Vehicle Clustering Method based on fuzzy C-means | |
CN110191148B (en) | Statistical function distributed execution method and system for edge calculation | |
CN106411770B (en) | A kind of data center network energy-saving routing algorithm based on SDN framework | |
CN111538587B (en) | Service function chain reconfiguration method based on load balancing | |
CN110519783B (en) | 5G network slice resource allocation method based on reinforcement learning | |
CN110087250B (en) | Network slice arranging scheme and method based on multi-objective joint optimization model | |
CN104619029B (en) | It is a kind of centralization cellular network architecture under baseband pool resource allocation methods and device | |
CN102664814A (en) | Grey-prediction-based adaptive dynamic resource allocation method for virtual network | |
CN109743217B (en) | Self-adaptive resource adjusting method based on SVRA algorithm | |
CN105813218A (en) | Service scheduling method of wireless Ad hoc network | |
US20220104127A1 (en) | Method and apparatus for power management in a wireless communication system | |
CN110198274A (en) | A kind of control method and its system of group flow | |
CN116389491B (en) | Cloud edge computing power resource self-adaptive computing system | |
CN104092620A (en) | Method and device for achieving adjustment of network bandwidth | |
CN106230722A (en) | SDN congestion link method of adjustment based on transfer value | |
Zhao et al. | PLOFR: An online flow route framework for power saving and load balance in SDN | |
CN112423041B (en) | Video stream processing method and system based on QoS constraint under distributed computing platform | |
CN114691372A (en) | Group intelligent control method of multimedia end edge cloud system | |
CN106850859A (en) | A kind of user's request distribution method of the CDN based on SDN | |
CN109862591A (en) | It is a kind of based on Qos eat dishes without rice or wine slice bandwidth borrow with caching sharing method | |
CN104579764A (en) | Dynamic data link resource allocation method based on shared time slot pools | |
CN116684418B (en) | Calculation power arrangement scheduling method, calculation power network and device based on calculation power service gateway | |
CN108965168A (en) | A kind of car networking based on utility function is dominant resource fairness distribution optimization method |
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 |