CN105430083A - Cloud platform deployment method - Google Patents
Cloud platform deployment method Download PDFInfo
- Publication number
- CN105430083A CN105430083A CN201510846477.6A CN201510846477A CN105430083A CN 105430083 A CN105430083 A CN 105430083A CN 201510846477 A CN201510846477 A CN 201510846477A CN 105430083 A CN105430083 A CN 105430083A
- Authority
- CN
- China
- Prior art keywords
- physical host
- computational resource
- virtual machine
- moment
- resource utilance
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
Abstract
The invention provides a cloud platform deployment method. The method comprises: judging whether a physical host is overloaded or not by calculating the utilization rate of a resource; and emigrating a part of virtual machines in the overloaded physical host. The cloud platform deployment method reduces the consumption of energy by comprehensive consideration, ensures the quality of service, and realizes the energy-saving goal of a large-sized cloud platform.
Description
Technical field
The present invention relates to cloud computing, particularly a kind of cloud Platform deployment method.
Background technology
Along with the fast development of cloud computing is with universal, the problem such as high energy consumption, high cost, poor efficiency of cloud platform becomes increasingly conspicuous.How while guarantee cloud computing reliable quality of service, optimizing cloud platform resource way to manage and power consumption mode, is the prerequisite building the sustainable development of cloud platform.In prior art, for realizing system-level energy consumption saving, virtual machine (vm) migration can improve cloud platform computational resource utilance and close idle physical host, brings pressure but to QoS and service level management, causes the combination property of cloud platform to decline.Therefore the efficient energy consumption for cloud platform reduces, and does not also find the optimum balance mode of the reduction of overall energy consumption and service quality.
Summary of the invention
For solving the problem existing for above-mentioned prior art, the present invention proposes a kind of cloud Platform deployment method, comprising:
Judge whether physical host transships by computational resource utilance;
A part of virtual machine in the physical host of overload is moved out.
Preferably, describedly judge whether physical host transships, and comprises further by computational resource utilance:
Statistics according to physical host computational resource utilance adjusts upper limit threshold, when the current computational resource utilance of physical host exceedes described upper limit threshold, determines that this physical host is in overload;
Every the computational resource utilance of a predetermined period record physical host, according to the resource utilization of physical host within cn cycle, obtain moment t
i-cn~ t
i-1computational resource compute utilization ratio statistics data be X
i-cn~ X
i-1, to moment t
icarry out following computing: R43=Q
3-Q
1, Q
3for X
i-cn~ X
i-1the numerical value of all numerical value ascending order arrangements the rear 75%th in sample, Q
1the numerical value of all numerical value ascending order arrangements the rear 25%th in this sample; By moment t
iupper limit threshold T
maxbe defined as T
max=1-S*R43; Wherein, adaptive threshold S needs to regulate according to system.
Preferably, at calculating moment t
iupper limit threshold T
maxin process, also comprise:
When estimating the computational resource utilance in next moment according to the computational resource compute utilization ratio statistics data in cn the cycle of physical host, obtain t
0~ t
cn-1the computational resource utilance sample data collection X that moment collects
0~ X
cn-1, estimate t
cnthe computational resource utilance X ' in moment
cn, sample X
ishared weight w
ibe defined as apart from the distance of estimation point according to sample
w
i(t)＝(1-((t
cn-1-t
i)-(t
cn-1-t
0))
3)
3
Estimate that current computational resource utilance is X '
cn=a+bt
cn, wherein (a, b) is calculated as follows:
Judge whether present physical main frame transships, or whether have virtual machine to need from present physical host migration, calculate whether meet S*X '
cn=S* (a+bt
cn)>=1, if meet, moves, otherwise then determines not transship.
Preferably, a part of virtual machine in the physical host of overload is moved out, comprises further:
The virtual machine of minimal time of selecting to move out needed on physical host carries out operation of moving out, until physical host disengaging overload, i.e. v ∈ V
n|
mEM (v)/LEFT
n≤ MEM (a)/LEFT
n, wherein V
nfor physical host Host
non virtual machine set, the memory source shared by virtual machine V is MEM (v), LEFT
nfor physical host Host
nremaining bandwidth resources.
The present invention compared to existing technology, has the following advantages:
The present invention proposes a kind of cloud Platform deployment method, consider and reduce energy consumption and guarantee service quality, realize the energy conservation object of large-scale cloud platform.
Accompanying drawing explanation
Fig. 1 is the flow chart of the cloud Platform deployment method according to the embodiment of the present invention.
Embodiment
Detailed description to one or more embodiment of the present invention is hereafter provided together with the accompanying drawing of the diagram principle of the invention.Describe the present invention in conjunction with such embodiment, but the invention is not restricted to any embodiment.Scope of the present invention is only defined by the claims, and the present invention contain many substitute, amendment and equivalent.Set forth many details in the following description to provide thorough understanding of the present invention.These details are provided for exemplary purposes, and also can realize the present invention according to claims without some in these details or all details.
An aspect of of the present present invention provides a kind of cloud Platform deployment method.Fig. 1 is the cloud Platform deployment method flow diagram according to the embodiment of the present invention.
The present invention determines mobility threshold adaptively, and the energy consumption realizing virtual machine (vm) migration under cloud platform reduces; On the other hand comprehensive modeling is carried out to energy consumption and service quality; not only have employed the virtual platform energy consumption comprehensive modeling mechanism that soft or hard combines; also service quality is quantized; seek virtual machine (vm) migration strategy best on multiple computing node, reduce target with the efficient energy consumption realizing large-scale cloud platform.
The event triggering virtual machine (vm) migration in the present invention mainly contains two classes.(1) physical host computational resource utilance exceedes most high threshold; (2) physical host computational resource utilance is lower than lowest threshold.The most frequently used physical host overload judgment mode is exactly manually arrange the most high threshold T of physical host computational resource utilance
max, more than T
maxi.e. predicate node overload, is alleviated node load by virtual machine on node of moving out, is improved the service quality of computing node.Consider that physical host computational resource utilance is a dynamic value, and have certain periodicity, fixed threshold is difficult to the real-time change adapting to cloud platform.The present invention is according to the statistics of physical host computational resource utilance, and the dynamic conditioning upper limit threshold of Corpus--based Method data adaptive, realizes the virtual machine adaptive-migration strategy under cloud platform.Meanwhile, when needs select virtual machine to move out from overload physical host, the suitable virtual machine of policy selection is adopted to implement to move out.
Automatically upper limit threshold is adjusted according to the computational resource compute utilization ratio statistics data on physical host.Every the computational resource utilance of a predetermined period record physical host, determine overload, if known t according to the resource utilization of physical host within cn cycle
i-cn~ t
i-1the computational resource compute utilization ratio statistics data in moment are X
i-cn~ X
i-1, following computing is carried out to moment ti: R43=Q
3-Q
1.In formula, Q
3for X
i-cn~ X
i-1the numerical value of all numerical value ascending order arrangements the rear 75%th in sample, Q
1the numerical value of all numerical value ascending order arrangements the rear 25%th in this sample.Moment t
itmax be defined as T
max=1-S*R43
In formula, adaptive threshold S can need to regulate according to system: S is lower, T
maxhigher, entire system energy consumption is lower; Otherwise it is poorer that energy consumption reduces effect.Cn on physical host computational resource compute utilization ratio statistics data are carried out ascending order arrangement by said method, then four parts are divided into, R43 is the amplitude of the 3rd part of last data and first part of last data difference, with this Computational Physics host calculating resource upper limit threshold, embody the dispersion degree of computational resource compute utilization ratio statistics value data.If computational resource compute utilization ratio statistics data are violent in certain phase change amplitude, the R43 in this period will be caused to raise, T
maxreduce, entire system energy consumption increases.Therefore, the present invention adopts following modification method to revise.
When needing to estimate the computational resource utilance in next moment according to the computational resource compute utilization ratio statistics data in cn the cycle of physical host, if t
0~ t
cn-1the computational resource utilance sample data that moment collects integrates as X
0~ X
cn-1, t
ithe computational resource utilance in moment is X
i, need to estimate t
cnthe computational resource utilance X ' in moment
cn, sample X
ishared weight w in fit procedure
iaccording to the distance to some extent difference of sample apart from estimation point, be defined as
w
i(t)＝(1-((t
cn-1-t
i)-(t
cn-1-t
0))
3)
3
Estimate that current computational resource utilance is X '
cn=a+bt
cn, wherein (a, b) is calculated as follows:
Judge whether present physical main frame transships, or whether have virtual machine to need from present physical host migration, calculate whether meet S*X '
cn=S* (a+bt
cn)>=1, if meet, moves, otherwise then determines not transship.
After judging whether overload by said process, need the partial virtual machine migration run by overload main frame so that no longer transship, the present invention adopts following strategy, and the virtual machine of minimal time of moving out needed for namely on selection physical host carries out operation of moving out, until physical host departs from overload.v∈V
n|
MEM(v)/LEFT
n≤MEM(a)/LEFT
n
Physical host Host
non virtual machine set expression be V
n, MEM (internal memory) resource shared by virtual machine V is MEM (v), Host
nremaining bandwidth resources are LEFT
n.Therefore, current hosts Host is selected
nthe minimum virtual machine of committed memory moves.
In sum, the present invention proposes a kind of cloud Platform deployment method, consider and reduce energy consumption and guarantee service quality, realize the energy conservation object of large-scale cloud platform.
Obviously, it should be appreciated by those skilled in the art, above-mentioned of the present invention each module or each step can realize with general computing system, they can concentrate on single computing system, or be distributed on network that multiple computing system forms, alternatively, they can realize with the executable program code of computing system, thus, they can be stored and be performed by computing system within the storage system.Like this, the present invention is not restricted to any specific hardware and software combination.
Should be understood that, above-mentioned embodiment of the present invention only for exemplary illustration or explain principle of the present invention, and is not construed as limiting the invention.Therefore, any amendment made when without departing from the spirit and scope of the present invention, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.In addition, claims of the present invention be intended to contain fall into claims scope and border or this scope and border equivalents in whole change and modification.
Claims (4)
1. a cloud Platform deployment method, is characterized in that, comprising:
Judge whether physical host transships by computational resource utilance;
A part of virtual machine in the physical host of overload is moved out.
2. method according to claim 1, is characterized in that, describedly judges whether physical host transships, and comprises further by computational resource utilance:
Statistics according to physical host computational resource utilance adjusts upper limit threshold, when the current computational resource utilance of physical host exceedes described upper limit threshold, determines that this physical host is in overload;
Every the computational resource utilance of a predetermined period record physical host, according to the resource utilization of physical host within cn cycle, obtain moment t
i-cn~ t
i-1computational resource compute utilization ratio statistics data be X
i-cn~ X
i-1, to moment t
icarry out following computing: R43=Q
3-Q
1, Q
3for X
i-cn~ X
i-1the numerical value of all numerical value ascending order arrangements the rear 75%th in sample, Q
1the numerical value of all numerical value ascending order arrangements the rear 25%th in this sample; By moment t
iupper limit threshold T
maxbe defined as T
max=1-S*R43; Wherein, adaptive threshold S needs to regulate according to system.
3. method according to claim 2, is characterized in that, at calculating moment t
iupper limit threshold T
maxin process, also comprise:
When estimating the computational resource utilance in next moment according to the computational resource compute utilization ratio statistics data in cn the cycle of physical host, obtain t
0~ t
cn-1the computational resource utilance sample data collection X that moment collects
0~ X
cn-1, estimate t
cnthe computational resource utilance X ' in moment
cn, sample X
ishared weight w
ibe defined as apart from the distance of estimation point according to sample
w
i(t)＝(1-((t
cn-1-t
i)-(t
cn-1-t
0))
3)
3
Estimate that current computational resource utilance is X '
cn=a+bt
cn, wherein (a, b) is calculated as follows:
Judge whether present physical main frame transships, or whether have virtual machine to need from present physical host migration, calculate whether meet S*X '
cn=S* (a+bt
cn)>=1, if meet, moves, otherwise then determines not transship.
4. method according to claim 3, is characterized in that, is moved out by a part of virtual machine in the physical host of overload, comprises further:
The virtual machine of minimal time of selecting to move out needed on physical host carries out operation of moving out, until physical host disengaging overload, namely
mEM (v)/LEFT
n≤ MEM (a)/LEFT
n, wherein V
nfor physical host Host
non virtual machine set, the memory source shared by virtual machine V is MEM (v), LEFT
nfor physical host Host
nremaining bandwidth resources.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510846477.6A CN105430083B (en) | 2015-11-27 | 2015-11-27 | A kind of cloud platform dispositions method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510846477.6A CN105430083B (en) | 2015-11-27 | 2015-11-27 | A kind of cloud platform dispositions method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105430083A true CN105430083A (en) | 2016-03-23 |
| CN105430083B CN105430083B (en) | 2018-12-04 |
Family
ID=55508019
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510846477.6A Active CN105430083B (en) | 2015-11-27 | 2015-11-27 | A kind of cloud platform dispositions method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105430083B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106598693A (en) * | 2016-05-11 | 2017-04-26 | 河南理工大学 | Energy consumption and load aware virtual machine integration method based on time delay strategy |
| CN108073443A (en) * | 2017-12-08 | 2018-05-25 | 中南大学 | Virtual machine selection based on the shared drive page and laying method in a kind of cloud data center |
| CN108196935A (en) * | 2017-12-06 | 2018-06-22 | 南京邮电大学 | A kind of energy saving moving method of virtual machine towards cloud computing |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102117226A (en) * | 2011-03-18 | 2011-07-06 | 浪潮(北京)电子信息产业有限公司 | Resource dispatching system and resource dispatching method |
| CN102185779A (en) * | 2011-05-11 | 2011-09-14 | 田文洪 | Method and device for realizing data center resource load balance in proportion to comprehensive allocation capability |
-
2015
- 2015-11-27 CN CN201510846477.6A patent/CN105430083B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102117226A (en) * | 2011-03-18 | 2011-07-06 | 浪潮(北京)电子信息产业有限公司 | Resource dispatching system and resource dispatching method |
| CN102185779A (en) * | 2011-05-11 | 2011-09-14 | 田文洪 | Method and device for realizing data center resource load balance in proportion to comprehensive allocation capability |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106598693A (en) * | 2016-05-11 | 2017-04-26 | 河南理工大学 | Energy consumption and load aware virtual machine integration method based on time delay strategy |
| CN108196935A (en) * | 2017-12-06 | 2018-06-22 | 南京邮电大学 | A kind of energy saving moving method of virtual machine towards cloud computing |
| CN108196935B (en) * | 2017-12-06 | 2021-11-02 | 南京邮电大学 | Cloud computing-oriented virtual machine energy-saving migration method |
| CN108073443A (en) * | 2017-12-08 | 2018-05-25 | 中南大学 | Virtual machine selection based on the shared drive page and laying method in a kind of cloud data center |
| CN108073443B (en) * | 2017-12-08 | 2021-10-26 | 中南大学 | Virtual machine selection and placement method based on shared memory page in cloud data center |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105430083B (en) | 2018-12-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xu et al. | A balanced virtual machine scheduling method for energy-performance trade-offs in cyber-physical cloud systems | |
| CN103365704B (en) | Internal memory pre-copy method in virtual machine (vm) migration and perform device and the system of the method | |
| CN103401938B (en) | Based on resource allocation system and the method thereof of traffic performance under distributed cloud framework | |
| US20160170469A1 (en) | Power balancing to increase workload density and improve energy efficiency | |
| CN113672383A (en) | Cloud computing resource scheduling method, system, terminal and storage medium | |
| CN105430083A (en) | Cloud platform deployment method | |
| CN104038392A (en) | Method for evaluating service quality of cloud computing resources | |
| CN104142860A (en) | Resource adjusting method and device of application service system | |
| Razali et al. | Virtual machine migration implementation in load balancing for cloud computing | |
| CN111290831A (en) | Virtual machine migration method based on reinforcement learning for cloud computing | |
| CN108228356A (en) | A kind of distributed dynamic processing method of flow data | |
| CN110875838A (en) | Resource deployment method, device and storage medium | |
| US9591067B2 (en) | Methods and apparatus for allocating cloud-based media resources | |
| US20140013142A1 (en) | Processing unit power management | |
| CN106201658A (en) | A kind of migration virtual machine destination host multiple-objection optimization system of selection | |
| CN104142850A (en) | Energy-saving scheduling method of data center | |
| CN107203256A (en) | Energy-conservation distribution method and device under a kind of network function virtualization scene | |
| CN110730232B (en) | High-energy-efficiency computing node selection and computing task allocation method in edge computing | |
| CN104978235A (en) | Operating frequency prediction based load balancing method | |
| CN110753101B (en) | Low-energy-consumption computing node selection and computing task allocation method in edge computing | |
| CN104580489A (en) | Web application elastic expansion method in cloud environment | |
| CN108073449B (en) | Dynamic virtual machine placement method | |
| CN108874517B (en) | Method for optimizing utilization rate division energy consumption of standby system with fixed priority | |
| CN105183537A (en) | Virtual machine migration processing method based on dynamic threshold window | |
| TWI646435B (en) | Serverless system and execution method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20181018 Address after: 200050 room 910-912, Tianshan Road, Changning District, Shanghai, 910-912 Applicant after: Craftsman advertising communication (Shanghai) Co., Ltd. Address before: 610041 No. 13, 2 building, 387 Tianren Road, high-tech zone, Chengdu, Sichuan, China 1307 Applicant before: CHENGDU WISHCLOUD TECHNOLOGY CO., LTD. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |