CN109062657A - Docker container dispatching method based on particle group optimizing - Google Patents
Docker container dispatching method based on particle group optimizing Download PDFInfo
- Publication number
- CN109062657A CN109062657A CN201810705833.6A CN201810705833A CN109062657A CN 109062657 A CN109062657 A CN 109062657A CN 201810705833 A CN201810705833 A CN 201810705833A CN 109062657 A CN109062657 A CN 109062657A
- Authority
- CN
- China
- Prior art keywords
- particle
- docker container
- node
- docker
- dispatching method
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- 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
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
-
- 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]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5077—Logical partitioning of resources; Management or configuration of virtualized resources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Complex Calculations (AREA)
Abstract
Present invention discloses a kind of Docker container dispatching method based on particle group optimizing, include the following steps: S1, particle coding step, particle swarm algorithm is applied in the scheduling of Docker container, it is by natural number coding, the particle code length in population is corresponding with the task of container;S2, initialization of population step, the position and speed of NP particle of random initializtion in problem solution space set particle swarm algorithm parameter;S3, fitness function step, using the quality of fitness function evaluation particle, using fitness function size as target, continuous iteration updates particle, until meeting stop condition, according to fitness function as a result, more new particle optimal value and global optimum;S4, simulated annealing step carry out simulated annealing to updated optimal value, particle swarm algorithm are avoided to fall into locally optimal solution.Dispatching method proposed by the invention not only realizes the load balancing of each node of Docker cluster, moreover it is possible to play the overall performance of cluster completely.
Description
Technical field
The present invention relates to a kind of dispatching methods, in particular to a kind of Docker container tune based on particle group optimizing
Degree method belongs to cloud computing technical field of virtualization.
Background technique
With the fast development of information technology and internet industry in recent years, cloud computing oneself through becoming current information technology
One of most important concept in field, referred to as the 4th of IT industry time revolution, becomes Future Internet and mobile Internet
The emerging calculating mode of the one kind combined.
Docker is that it is virtual to belong to operating system layer based on an application container engine of linux container (LXC) creation
Change, is mainly used for the problem of settlement server application rapid build, deployment are with sharing.Docker has been quickly grown since birth,
The (SuSE) Linux OS of present mainstream has all supported Docker, at the same the engine also obtained include Google, Microsoft, IBM,
The support energetically of numerous enterprises including Amazon and VMware, and obtained in the cloud platform and service product of these companies
Widely apply.
Application developer Docker can be used to be packaged application and run required dependence packet to a transplantable appearance
In device, then it is published on any machine for supporting Docker, to realize the virtualization of lightweight.The foundation stone of cloud computing at present
It is the other isolation of operating system grade, fictionalizes multiple main frames on same physical server.The appearance of Docker, actually
A kind of other isolation of application-level is completed, it changes exploitation basic at present, operating unit, by directly operating fictitious host computer
(VM), it is transformed on the container of operation sequence operation.
In addition, the appearance of Docker cluster management instrument Swarm further promotes the use of Docker in the cluster, it is
Cloud computing platform and the virtual method of data center provide a kind of new thinking.
Current Docker container scheduling is based primarily upon two ways, and one is traditional dispatching method, another kind is to open
Hairdo dispatching algorithm.How both above-mentioned modes are subjected to combination appropriate, the load that each node of cluster is better achieved is equal
Weighing apparatus, the overall performance for playing cluster also just become to fully embody the advantage using lightweight virtualization Docker
Those skilled in that art institute urgent problem to be solved at present.
Summary of the invention
In view of the prior art, there are drawbacks described above, and the purpose of the present invention is to propose to a kind of Docker based on particle group optimizing
Container dispatching method.
A kind of Docker container dispatching method based on particle group optimizing, includes the following steps:
S1, particle coding step, particle swarm algorithm are applied in the scheduling of Docker container, by natural number coding,
Particle code length in population is corresponding with the task of container;
S2, initialization of population step, the position and speed of NP particle of random initializtion in problem solution space set grain
Swarm optimization parameter;
S3, fitness function step, using the quality of fitness function evaluation particle, particle is with fitness function size
Target, continuous iteration updates, until meeting stop condition, then according to fitness function as a result, updating the grain of each particle
The global optimum of sub- optimal value and all particles;
S4, simulated annealing step carry out simulated annealing to updated particle optimal value and global optimum, avoid particle
Group's algorithm falls into locally optimal solution.
Preferably, the S1 particle coding step, including operate as follows:
Equipped with m task, a task corresponds to a Docker container, has n node resource in Docker cluster, then
It is vector that particle, which may be encoded as n, and expression formula is,
P={ p1,p2,…pi…,pm,
Wherein, 1≤pi≤ n, every one-dimensional coordinate of particle indicate the number of a Docker container, and each Docker holds
Device has a task, any one-dimensional component pmThe resource number of this container is distributed in expression.
Preferably, the S2 initialization of population step, including operate as follows:
If population scale is NP, m task is scheduled on n resource node, system is random when initialization of population
NP particle is generated, the position of each particle is by vector P expression, the expression formula of i-th of particle,
pi={ pi1,pi2,…,pij,…,pim,
Wherein, 1≤pij≤ n indicates that task j is assigned to pthijOn number node, pijIt is initialized as random whole between (1, n)
Number,
The speed of i-th of particle is expressed as by vector v,
vi={ vi1,vi2,…,vij,…,vim,
Wherein, 1≤i≤NP ,-n≤vij≤ n, vijThe random number being initialized as between (- n, n).
Preferably, the S3 fitness function step, including following sub-step:
S31, service-level agreement is evaluated using the CPU usage of node;
S32, resource utilization is evaluated using surplus resources utilization rate;
S33, corresponding weight is set according to the requirement to objective optimization, obtains comprehensive fitness degree function.
Preferably, the expression formula of the service-level agreement evaluation function used in S31 based on CPU usage are as follows:
Wherein, UcpuFor the CPU usage of node, p is the threshold range that regulation guarantees service-level agreement.
Preferably, the expression formula of surplus resources utilization rate function used in S32 are as follows:
Wherein, RiFor i-th dimension surplus resources, RminFor the minimum value of surplus resources.
Preferably, the expression formula of comprehensive fitness degree function obtained in S33 are as follows:
f(Ucpu,Umem)=K1fsLA+K2fr,
Wherein, UcpuFor the CPU usage of node, UmemFor the memory usage of node, K1、K2For weight.
Preferably, in the S4 simulated annealing step, the expression formula of annealing process are as follows:
T (t+1)=α × T (t),
Wherein, α is cooling decay factor in annealing simulation, and T is temperature, and t is control parameter.
Compared with prior art, advantages of the present invention is mainly reflected in the following aspects:
The present invention completes the scheduling of Docker container using the particle swarm algorithm optimized, will be appointed by natural number coding
Business is placed on corresponding resource node, while possessing which task, whole process logic using decoding to find resource node
Clearly, significant effect.Meanwhile simulated annealing thought is employed herein, it prevents particle swarm algorithm from falling into locally optimal solution, has
Improve to effect the accuracy of the ability of searching optimum, efficiency and optimal solution of population.In addition, the present invention is also in same domain
Other relevant issues provide reference, can carry out expansion extension on this basis, apply to other dispatching methods in field
In technical solution, there is very wide application prospect.
In general, dispatching method proposed by the invention not only realize each node of Docker cluster load it is equal
Weighing apparatus, moreover it is possible to play the overall performance of cluster completely, to fully play the advantage of Docker container lightweight, have very high
Use and promotional value.
Just attached drawing in conjunction with the embodiments below, the embodiment of the present invention is described in further detail, so that of the invention
Technical solution is more readily understood, grasps.
Detailed description of the invention
Fig. 1 is algorithm flow chart of the invention;
Fig. 2 is the comparison diagram of virtual machine and Docker container;
Fig. 3 is will to optimize particle swarm algorithm and be fused in Docker container to scheme;
Fig. 4 is container task schedule figure in Docker cluster;
Fig. 5 is the performance map of container in Docker cluster.
Specific embodiment
As shown in fig. 1~fig. 5, present invention discloses a kind of Docker container dispatching method based on particle group optimizing, packet
Include following steps:
S1, particle coding step, particle swarm algorithm are applied in the scheduling of Docker container, by natural number coding,
Particle code length in population is corresponding with the task of container;
S2, initialization of population step, the position and speed of NP particle of random initializtion in problem solution space set grain
Swarm optimization parameter;
S3, fitness function step, using the quality of fitness function evaluation particle, particle is with fitness function size
Target, continuous iteration updates, until meeting stop condition, then according to fitness function as a result, updating the grain of each particle
The global optimum of sub- optimal value and all particles;
S4, simulated annealing step carry out simulated annealing to updated particle optimal value and global optimum, avoid particle
Group's algorithm falls into locally optimal solution.
Specifically, the S1 particle coding step, comprising:
Particle swarm algorithm is applied in the scheduling of Docker container, since task can only take in the scheduling of Docker container
Discrete value, the coding to particle also can only be discrete value.So first having to encode particle, by the position of particle, speed
Degree combines with Docker scheduler task.Natural number coding is used in the present embodiment, allows the code length and container of particle
Task it is corresponding.
Equipped with m task, a task corresponds to a Docker container, has n node resource in Docker cluster, then
It is vector that particle, which may be encoded as n, and expression formula is,
P={ p1,p2,…pi…,pm,
Wherein, 1≤pi≤ n, every one-dimensional coordinate of particle indicate the number of a Docker container, and each Docker holds
Device has a task, any one-dimensional component pmThe resource number of this container is distributed in expression.
For example, when task number is m=12, when Docker node resource number n=5, i.e. 12 tasks distribution
Onto 5 node resources, particle may be encoded as (5,4,3,2,3,2,4,2,4,2, Isosorbide-5-Nitrae), as shown in table 1.Decoding to particle
It may know that task distribution condition, as shown in table 2.Wherein subtask 7,8 is assigned in No. 1 resource, and subtask 2,6 is assigned to 2
In number resource, subtask 1,5,10,11 is assigned in No. 3 resources, and subtask 4 is assigned in No. 4 resources, 3,9,12 points of subtask
It is fitted in No. 5 resources.
1 particle encoding examples of table
Task number | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
Resource number | 5 | 4 | 3 | 2 | 3 | 2 | 4 | 2 | 4 | 2 | 1 | 4 |
2 particle of table decodes numbering example
Resource number | 1 | 2 | 3 | 4 | 5 |
The task number of distribution | 11 | 4 6 8 10 | 3 5 | 2 7 9 12 | 1 |
The S2 initialization of population step, comprising:
If population scale is NP, m task is scheduled on n resource node, system is random when initialization of population
NP particle is generated, the position of each particle is by vector P expression, the expression formula of i-th of particle,
pi={ pi1,pi2,…,pij,…,pim,
Wherein, 1≤pij≤ n indicates that task j is assigned to pthijOn number node, pijIt is initialized as random whole between (1, n)
Number,
The speed of i-th of particle is expressed as by vector v,
vi={ vi1,vi2,…,vij,…,vim,
Wherein, 1≤i≤NP ,-n≤vij≤ n, vijThe random number being initialized as between (- n, n).
The S3 fitness function step, including following sub-step:
S31, service-level agreement is evaluated using the CPU usage of node, it is used based on CPU usage
The expression formula of service-level agreement evaluation function are as follows:
Wherein, UcpuFor the CPU usage of node, p is the threshold range that regulation guarantees service-level agreement.
S32, resource utilization is evaluated using surplus resources utilization rate, used surplus resources utilization rate function
Expression formula are as follows:
Wherein, RiFor i-th dimension surplus resources, RminFor the minimum value of surplus resources.
S33, corresponding weight is set according to the requirement to objective optimization, obtains comprehensive fitness degree function, it is obtained comprehensive
Close the expression formula of fitness function are as follows:
f(Ucpu,Umem)=K1fSLA+K2fr,
Wherein, UcpuFor the CPU usage of node, UmemFor the memory usage of node, K1、K2For weight.
In the S4 simulated annealing step, the expression formula of annealing process are as follows:
T (t+1)=α × T (t),
Wherein, α is cooling decay factor in annealing simulation, generally take be slightly less than 1.00 normal number, T is temperature, and t is control
Parameter processed.
The fitness value of particle is calculated by fitness function after obtaining new position for each particle, if suitable
It answers angle value to be better than the position of current particle, then new position is moved the particles to, if the fitness value calculated does not have
Better than current location, then new position is moved to certain probability.According to simulated annealing thought, there is each particle in population
One has respective annealing process, and the position that each particle updates oneself is also selected with probability, to control particle
Update position, avoid the case where falling into locally optimal solution.
It is combined in conclusion the present invention passes through the particle swarm algorithm optimized with Docker container, passes through nature first
Task is assigned on each resource node by number encoder, that is, a kind of mapping is formed between container and node, will be had and be appointed
In the container mappings to node of business, and find which task resource node possesses using decoding.It is carried out secondly by population
Initialization, m task is assigned on n node, then by fitness function, to the grain of each particle and entire population
The speed and position of son are updated, and find the highest node of resource utilization.Finally by simulated annealing thought, particle is prevented
Locally optimal solution is fallen into during update, improves the update efficiency and accuracy of particle.Finally to implementation of the invention
Property verified, the results show that the present invention has significantly in resource allocation and performance compared to the dispatching algorithm that Docker is carried
It is promoted.
The present invention also provides reference for other relevant issues in same domain, can carry out expanding on this basis and prolong
It stretches, applies in field in the technical solution of other dispatching methods, there is very wide application prospect.
In general, dispatching method proposed by the invention not only realize each node of Docker cluster load it is equal
Weighing apparatus, moreover it is possible to play the overall performance of cluster completely, to fully play the advantage of Docker container lightweight, have very high
Use and promotional value.
It is obvious to a person skilled in the art that invention is not limited to the details of the above exemplary embodiments, Er Qie
In the case where without departing substantially from spirit and essential characteristics of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Benefit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent elements of the claims
Variation is included within the present invention, and any reference signs in the claims should not be construed as limiting the involved claims.
In addition, it should be understood that although this specification is described in terms of embodiments, but not each embodiment is only wrapped
Containing an independent technical solution, this description of the specification is merely for the sake of clarity, and those skilled in the art should
It considers the specification as a whole, the technical solutions in the various embodiments may also be suitably combined, forms those skilled in the art
The other embodiments being understood that.
Claims (8)
1. a kind of Docker container dispatching method based on particle group optimizing, which comprises the steps of:
S1, particle coding step, particle swarm algorithm are applied in the scheduling of Docker container, by natural number coding, by grain
Particle code length in subgroup is corresponding with the task of container;
S2, initialization of population step, the position and speed of NP particle of random initializtion in problem solution space set population
Algorithm parameter;
S3, fitness function step, using the quality of fitness function evaluation particle, particle is using fitness function size as mesh
Mark, continuous iteration updates, until meeting stop condition, then according to fitness function as a result, updating the particle of each particle
The global optimum of optimal value and all particles;
S4, simulated annealing step carry out simulated annealing to updated particle optimal value and global optimum, population are avoided to calculate
Method falls into locally optimal solution.
2. the Docker container dispatching method according to claim 1 based on particle group optimizing, which is characterized in that the S1
Particle coding step, including operate as follows:
Equipped with m task, a task corresponds to a Docker container, has n node resource in Docker cluster, then particle
May be encoded as n is vector, and expression formula is,
P={ p1,p2,…pi…,pm,
Wherein, 1≤pi≤ n, every one-dimensional coordinate of particle indicate the number of a Docker container, and each Docker container is equal
There is a task, any one-dimensional component pmThe resource number of this container is distributed in expression.
3. the Docker container dispatching method according to claim 1 based on particle group optimizing, which is characterized in that the S2
Initialization of population step, including operate as follows:
If population scale is NP, m task is scheduled on n resource node, system is randomly generated when initialization of population
NP particle, the position of each particle indicate by vector P, and the expression formula of i-th of particle is,
pi={ pi1,pi2,…,pij,…,pim,
Wherein, 1≤pij≤ n indicates that task j is assigned to pthijOn number node, pijThe random integers being initialized as between (1, n),
The speed of i-th of particle is expressed as by vector v,
vi={ vi1,vi2,…,vij,…,vim,
Wherein, 1≤i≤NP ,-n≤vij≤ n, vijThe random number being initialized as between (- n, n).
4. the Docker container dispatching method according to claim 1 based on particle group optimizing, which is characterized in that the S3
Fitness function step, including following sub-step:
S31, service-level agreement is evaluated using the CPU usage of node;
S32, resource utilization is evaluated using surplus resources utilization rate;
S33, corresponding weight is set according to the requirement to objective optimization, obtains comprehensive fitness degree function.
5. the Docker container dispatching method according to claim 4 based on particle group optimizing, which is characterized in that in S31
The expression formula of the used service-level agreement evaluation function based on CPU usage are as follows:
Wherein, UcpuFor the CPU usage of node, p is the threshold range that regulation guarantees service-level agreement.
6. the Docker container dispatching method according to claim 4 based on particle group optimizing, which is characterized in that in S32
The expression formula of used surplus resources utilization rate function are as follows:
Wherein, RiFor i-th dimension surplus resources, RminFor the minimum value of surplus resources.
7. the Docker container dispatching method according to claim 4 based on particle group optimizing, which is characterized in that in S33
The expression formula of comprehensive fitness degree function obtained are as follows:
f(Ucpu,Umem)=K1fSLA+K2fr,
Wherein, UcpuFor the CPU usage of node, UmemFor the memory usage of node, K1、K2For weight.
8. the Docker container dispatching method according to claim 1 based on particle group optimizing, which is characterized in that described
In S4 simulated annealing step, the expression formula of annealing process are as follows:
T (t+1)=α × T (t),
Wherein, α is cooling decay factor in annealing simulation, and T is temperature, and t is control parameter.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810705833.6A CN109062657A (en) | 2018-07-02 | 2018-07-02 | Docker container dispatching method based on particle group optimizing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810705833.6A CN109062657A (en) | 2018-07-02 | 2018-07-02 | Docker container dispatching method based on particle group optimizing |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109062657A true CN109062657A (en) | 2018-12-21 |
Family
ID=64818192
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810705833.6A Pending CN109062657A (en) | 2018-07-02 | 2018-07-02 | Docker container dispatching method based on particle group optimizing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109062657A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110058924A (en) * | 2019-04-23 | 2019-07-26 | 东华大学 | A kind of container dispatching method of multiple-objection optimization |
CN111858029A (en) * | 2020-06-16 | 2020-10-30 | 国网福建省电力有限公司信息通信分公司 | Storm cluster load balancing method and system based on discrete particle swarm |
CN114579294A (en) * | 2020-12-02 | 2022-06-03 | 上海交通大学 | Container elastic telescopic system supporting service load surge prediction under cloud native environment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104331749A (en) * | 2014-10-24 | 2015-02-04 | 陕西科技大学 | AGV optimization scheduling method based on simulated annealing particle swarm |
CN107491341A (en) * | 2017-08-31 | 2017-12-19 | 福州大学 | A kind of virtual machine distribution method based on particle group optimizing |
CN107766135A (en) * | 2017-09-29 | 2018-03-06 | 东南大学 | Method for allocating tasks based on population and simulated annealing optimization in mobile cloudlet |
CN108182115A (en) * | 2017-12-28 | 2018-06-19 | 福州大学 | A kind of virtual machine load-balancing method under cloud environment |
-
2018
- 2018-07-02 CN CN201810705833.6A patent/CN109062657A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104331749A (en) * | 2014-10-24 | 2015-02-04 | 陕西科技大学 | AGV optimization scheduling method based on simulated annealing particle swarm |
CN107491341A (en) * | 2017-08-31 | 2017-12-19 | 福州大学 | A kind of virtual machine distribution method based on particle group optimizing |
CN107766135A (en) * | 2017-09-29 | 2018-03-06 | 东南大学 | Method for allocating tasks based on population and simulated annealing optimization in mobile cloudlet |
CN108182115A (en) * | 2017-12-28 | 2018-06-19 | 福州大学 | A kind of virtual machine load-balancing method under cloud environment |
Non-Patent Citations (3)
Title |
---|
CHUNXIAO FAN等: "Research on Improved 2D-BPSO-based VMContainer Hybrid Hierarchical Cloud Resource Scheduling Mechanism", 《2016 IEEE INTERNATIONAL CONFERENCE ON COMPUTER AND INFORMATION TECHNOLOGY (CIT)》 * |
娄建峰等: "基于改进粒子群算法的云计算任务调度算法", 《微电子学与计算机》 * |
王德文等: "基于改进粒子群算法的云计算平台资源调度", 《计算机应用研究》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110058924A (en) * | 2019-04-23 | 2019-07-26 | 东华大学 | A kind of container dispatching method of multiple-objection optimization |
CN110058924B (en) * | 2019-04-23 | 2023-08-04 | 东华大学 | Multi-objective optimized container scheduling method |
CN111858029A (en) * | 2020-06-16 | 2020-10-30 | 国网福建省电力有限公司信息通信分公司 | Storm cluster load balancing method and system based on discrete particle swarm |
CN111858029B (en) * | 2020-06-16 | 2023-06-27 | 国网福建省电力有限公司信息通信分公司 | Storm cluster load balancing method and system based on discrete particle swarm |
CN114579294A (en) * | 2020-12-02 | 2022-06-03 | 上海交通大学 | Container elastic telescopic system supporting service load surge prediction under cloud native environment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhao et al. | Offloading tasks with dependency and service caching in mobile edge computing | |
Prem Jacob et al. | A multi-objective optimal task scheduling in cloud environment using cuckoo particle swarm optimization | |
Almezeini et al. | Task scheduling in cloud computing using lion optimization algorithm | |
Li et al. | Energy-efficient migration and consolidation algorithm of virtual machines in data centers for cloud computing | |
CN110532092A (en) | Reso urce matching method, device and equipment | |
Abdulhamid et al. | Secure scientific applications scheduling technique for cloud computing environment using global league championship algorithm | |
Shen et al. | Adaptive task scheduling strategy in cloud: when energy consumption meets performance guarantee | |
CN109062657A (en) | Docker container dispatching method based on particle group optimizing | |
CN103970609A (en) | Cloud data center task scheduling method based on improved ant colony algorithm | |
CN103281374B (en) | A kind of method of data fast dispatch during cloud stores | |
Al-Olimat et al. | Cloudlet scheduling with particle swarm optimization | |
CN109067834B (en) | Discrete particle swarm scheduling algorithm based on oscillation type inertia weight | |
Wang et al. | DCCP: an effective data placement strategy for data-intensive computations in distributed cloud computing systems | |
Ababneh | A hybrid approach based on grey wolf and whale optimization algorithms for solving cloud task scheduling problem | |
Lou et al. | Efficient container assignment and layer sequencing in edge computing | |
Li et al. | A particle swarm optimization-based container scheduling algorithm of docker platform | |
Wang et al. | Task scheduling algorithm based on improved Min-Min algorithm in cloud computing environment | |
Yousefipour et al. | Improving the load balancing and dynamic placement of virtual machines in cloud computing using particle swarm optimization algorithm | |
Yakubu et al. | An efficient meta-heuristic resource allocation with load balancing in IoT-Fog-cloud computing environment | |
Liu et al. | Energy‐aware task scheduling with time constraint for heterogeneous cloud datacenters | |
WO2022015322A1 (en) | Concurrent hash map updates | |
Kumar et al. | QoS‐aware resource scheduling using whale optimization algorithm for microservice applications | |
Gong et al. | Task scheduling in cloud computing environment based on enhanced marine predator algorithm | |
Bhagavathi et al. | Improved beetle swarm optimization algorithm for energy efficient virtual machine consolidation on cloud environment | |
Abd Elaziz et al. | Hybrid enhanced optimization-based intelligent task scheduling for sustainable edge computing |
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 |
Application publication date: 20181221 |
|
RJ01 | Rejection of invention patent application after publication |