WO2016165392A1 - Genetic algorithm-based cloud computing resource scheduling method - Google Patents
Genetic algorithm-based cloud computing resource scheduling method Download PDFInfo
- Publication number
- WO2016165392A1 WO2016165392A1 PCT/CN2015/098620 CN2015098620W WO2016165392A1 WO 2016165392 A1 WO2016165392 A1 WO 2016165392A1 CN 2015098620 W CN2015098620 W CN 2015098620W WO 2016165392 A1 WO2016165392 A1 WO 2016165392A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- chromosome
- host
- population
- cloud computing
- subtask
- Prior art date
Links
Images
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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L65/00—Network arrangements, protocols or services for supporting real-time applications in data packet communication
- H04L65/40—Support for services or applications
Definitions
- the invention belongs to the field of cloud computing, and particularly relates to a cloud computing resource scheduling method based on a genetic algorithm.
- Cloud computing As a new type of information processing, has penetrated into all areas of our life and work. Cloud computing is supported by virtualization technology. The most basic idea is to use it on demand. In this process, because the energy consumption of cloud data center and its resource provision efficiency become the key issues affecting cloud computing performance, how effective is it. Reasonable use of cloud computing resources has also become a key point.
- Cloud computing is a further development of distributed computing, parallel processing and grid computing. It is an Internet-based computing system that provides hardware services, infrastructure services, platform services, software services, and storage services to various Internet applications.
- Cloud computing is a convenient, on-demand access to a configurable computing resource pool (including networks) over the network.
- cloud computing model has 5 basics Features: On-demand self-service, extensive network access, shared resource pools, rapid resiliency, measurable services, and three service models: Software as a Service (SaaS), Platform as a Service (PaaS), Infrastructure Services (IaaS), and four deployment methods: private cloud, community cloud, public cloud, hybrid cloud.
- SaaS Software as a Service
- PaaS Platform as a Service
- IaaS Infrastructure Services
- cloud computing The service provided by cloud computing is huge, so the number of tasks in the "cloud” is huge.
- the system comes out with a large number of people every moment, so resource management is a key issue in cloud computing.
- Its scheduling strategy and algorithm directly affect the performance and cost of the cloud system.
- cloud data centers are becoming larger and larger, and their energy consumption problems are becoming more serious.
- the global data center increased by about 56% between 2005 and 2010, and the energy consumption of the US data center increased by 36%.
- a 50,000-node data center consumes more than 100 million kilowatts of electricity per year, accounting for 40% of a data center's operating costs.
- the data center of communication operators is a large power consumer.
- China Telecom's data center consumes 1.12 billion kWh per year.
- China Unicom's data center is 990 million kWh by 2020.
- the energy consumption of major cloud computing operators around the world will be close. 2 trillion kWh. Therefore, there is an urgent need to study how to effectively use resources in the cloud computing environment.
- the object of the present invention is to overcome the shortcomings and shortcomings of the prior art, and to provide a cloud computing resource scheduling method based on a genetic algorithm.
- the method can make the scheduling of cloud computing resources in real time, determine the best resource construction and adjustment strategy, and improve the utility ratio of cloud computing under the premise of guaranteeing the service quality of the cloud computing network.
- a cloud computing resource scheduling method based on a genetic algorithm the steps are as follows:
- the cloud task submitted by the user is divided into several subtasks, and an expected Qos (Quality of Service) is set according to the cloud task;
- the expected Qos refers to the time and cost of the user when acquiring the cloud computing resource.
- initialization initialize the population size, and set the maximum number of iterations of the population
- Generating an initial population randomly generating a chromosome according to an encoding rule, abstracting a cloud task into a coding input of a chromosome, each gene in the chromosome is a subtask in the cloud task, and a bit number of the chromosome indicates that each subtask representative is assigned to a virtual The number of the machine; then determine whether the chromosome satisfies the expected Qos, and if so, add the chromosome to the initial population, and if not, discard the chromosome until the size of the initial population is reached;
- selection, crossover, and mutation operations selecting chromosomes with high fitness values according to selection probabilities, copying these chromosomes to the next generation of new populations; and performing crossover and mutation operations on the remaining chromosomes in the current population;
- step S6 Determine whether the chromosome after the crossover and mutation operation in step S5 satisfies the expected QoS, if satisfied, Then join the next generation of new populations, if not satisfied, discard until the new population reaches the population size, and the number of population iterations is increased by 1;
- step S7 it is determined whether the number of population iterations reaches the maximum number of iterations, if not, then returns to step S4, and if so, proceeds to step S8;
- chromosome coding is performed by using a path-based coding method.
- step S4 the fitness of each chromosome in the population is calculated by the following adaptive function:
- EC total is the total energy consumption of the resource scheduling scheme in chromosome X k
- f(X k ) is the fitness of chromosome X k
- the total energy consumption EC total of the resource scheduling scheme in the chromosome X k is:
- WEC j is the working energy consumption of the host j in one cycle
- m j is the number of subtasks running on the host j
- DEC j is the energy consumption of the host j in the dormant state in one cycle
- IEC j is the sink The energy consumption when host j is idle
- N is the number of hosts
- CT ij represents the power consumption of subtask i on host j
- CI i represents the number of execution commands of subtask i on host j
- CS j represents the CPU processing speed of host j
- PC j represents the operation of host j Power consumption
- SEC ij represents the storage energy consumption of subtask i on host j
- SD i represents the amount of data that subtask i needs to read and write on host j
- SS j represents the disk read rate of host j
- PS j represents host j Store power consumption.
- the selection probability in the selection operation in the step S5 is a roulette operator, and the selection probability of each chromosome in the population is:
- popSize is the size of the population.
- the mutation operation in the step S5 adopts a substitution variation method, first selecting a sub-bit string from the parent chromosome, and then randomly selecting a position in the remaining bit string and inserting the sub-bit string.
- the mutation operation in the step S5 adopts a partial matching method, and two intersection points are randomly selected, the position between the two points will be crossed, and the other positions are copied.
- the execution time of each subtask in each host satisfies the following conditions:
- TE ij CT ij +ST ij ;
- TE ij represents the total running time of the subtask i on the host j
- CT ij represents the task execution time of the subtask i on the host j
- ST ij represents the time of reading and writing data on the host j
- Host j in the execution time of all subtasks T j satisfies the following conditions:
- m j is the number of subtasks in the host j
- N is the number of hosts
- the expected cost in the Qos set in the step S1 satisfies the following conditions:
- cost ij represents the cost of executing subtask i in host j
- s Re q i represents the resource corresponding to subtask i of host j
- s Pr ice j represents the resource corresponding to subtask i in host j The cost.
- the MapReduce programming model in the cloud computing is used to divide the cloud task submitted by the user into a plurality of subtasks.
- the method of the present invention uses a genetic algorithm to obtain an optimal resource scheduling scheme in the cloud computing.
- the expected QoS is set according to the cloud task, wherein the expected QoS takes time for the user to acquire the cloud computing resource.
- the chromosomes that satisfy the expected Qos are left in the iterative update of the population, the chromosomes that do not satisfy the expected Qos are discarded, and will be adapted according to the selection probability
- the chromosome with high degree of value is inherited into the next generation population, which ensures the quality of the chromosome in the population, realizes the on-demand use and flexible expansion of the resources of cloud computing, and can make the scheduling of cloud computing resources in real time, making the best decision.
- the resource construction and adjustment strategy improves the utility ratio of cloud computing on the premise of guaranteeing the quality of cloud computing network services.
- the fitness of the chromosome is calculated by an adaptive function related to the total energy consumption of the resource scheduling scheme in the chromosome, and the fitness function can evaluate the pros and cons of the chromosome, and the larger the fitness value, the chromosome The better the fitness, and vice versa, the more accurate the chromosomes that meet the actual optimal resource scheduling requirements can be selected by the fitness function, which further ensures that the resource scheduling scheme finally obtained by the genetic algorithm is the optimal solution.
- Figure 1 is a flow chart of the method of the present invention.
- this embodiment discloses a cloud computing resource scheduling method based on a genetic algorithm, and the steps are as follows:
- the cloud task submitted by the user is divided into several subtasks, and an expected Qos is set according to the cloud task; the expected Qos refers to the time and cost of the user when acquiring the cloud computing resource.
- a request Using the MapReduce programming model of cloud computing, the cloud task submitted by the user is divided into several subtasks, and an expected Qos is set according to the cloud task; the expected Qos refers to the time and cost of the user when acquiring the cloud computing resource. A request.
- initialization Initialize the population size popSize, and set the population maximum iteration number Tmax.
- Generating an initial population randomly generating a chromosome according to an encoding rule, abstracting the cloud task into a coding input of the chromosome, each gene in the chromosome is a subtask in the cloud task, and the bit number of the chromosome indicates that each subtask is assigned to the virtual machine. The number is then determined whether the chromosome satisfies the expected QoS, and if so, the chromosome is added to the initial population, and if not, the chromosome is discarded until the size of the initial population is reached.
- the path-based coding method is used to perform coding of a chromosome, and of course, a coding method based on binary, matrix-based, contiguous, index-based, etc. may be employed.
- EC total is the total energy consumption of the resource scheduling scheme in chromosome X k ; the fitness function evaluates the pros and cons of the chromosome, and the greater the fitness value, the better the fitness of the chromosome, and vice versa. The smaller the EC total in this step, the larger the fitness value.
- the total energy consumption EC total of the resource scheduling scheme in chromosome X k is:
- WEC j is the working energy consumption of the host j in one cycle
- m j is the number of subtasks running on the host j
- DEC j is the energy consumption of the host j in the dormant state in one cycle
- IEC j is the sink The energy consumption when host j is idle
- N is the number of hosts
- CT ij represents the power consumption of subtask i on host j
- CI i represents the number of execution commands of subtask i on host j
- CS j represents the CPU processing speed of host j
- PC j represents the operation of host j Power consumption
- SEC ij represents the storage energy consumption of subtask i on host j
- SD i represents the amount of data that subtask i needs to read and write on host j
- SS j represents the disk read rate of host j
- PS j represents host j Store power consumption.
- selection, crossover and mutation operations selecting chromosomes with high fitness values according to the selection probability, copying these highly adaptive chromosomes to the next generation of new populations; and performing crossover and mutation operations on the remaining chromosomes in the current population.
- the selection probability in the selection operation uses the roulette operator, and the probability that the chromosome in the population is selected is proportional to its fitness; the selection probability of each chromosome in the population is:
- popSize is the size of the population and f(X k ) is the fitness of the chromosome X k .
- the substitution variation method is adopted, first selecting a sub-bit string from the parent chromosome, and then randomly selecting a position in the remaining bit string and inserting the sub-bit string.
- the mutation operation in this embodiment may also adopt exchange mutation (EM), insertion mutation (IM), and simple inversion variation (SIM).
- EM exchange mutation
- IM insertion mutation
- SIM simple inversion variation
- IVM inversion variation
- SM competition for variation
- the mutation operation adopts a partial matching method, and two intersection points are randomly selected, and the position between the two points will be crossed, and the other positions are copied.
- the cross operation in this embodiment may also adopt a partial matching method (PMX), a cyclic cross method (CX), an order cross method (OX), a position based cross method (POS), and the like.
- step S6 Determine whether the chromosome after the crossover and mutation operation in step S5 satisfies the expected QoS, and if yes, join the next generation new population, and if not, discard until the new population reaches the population size popSize, and the number of population iterations T plus 1.
- step S7 Determine whether the number of population iterations reaches the maximum number of iterations Tmax. If not, return to step S4, and if yes, proceed to step S8.
- Cloud computing Qos means that when consumers get cloud computing resources, they hope that time, cost and other indicators can meet the expected goals of consumers.
- consumers rely on resource providers to meet their computing needs, in order to match their requirements, the multi-dimensional Qos requirements (time, cost) of the user task must be guaranteed.
- the energy consumption of material resources is equal to the product of power consumption and time.
- the power consumption of resources in different situations is also different.
- Physical hosts are usually divided into three states: sleep, idle, and work.
- the execution time of each subtask in each host satisfies the following conditions:
- TE ij CT ij +ST ij ;
- TE ij represents the total running time of the subtask i on the host j
- CT ij represents the task execution time of the subtask i on the host j
- ST ij represents the time of reading and writing data on the host j
- Host j in the execution time of all subtasks T j satisfies the following conditions:
- m j is the number of subtasks in the host j
- N is the number of hosts
- the expected cost in the Qos set in the step S1 satisfies the following conditions:
- cost ij represents the cost of executing subtask i in host j
- s Re q i represents the resource corresponding to subtask i of host j
- s Pr ice j represents the resource corresponding to subtask i in host j The cost.
- the resource scheduling solution is a trade-off between time and cost.
- an expected time and cost is obtained as the QoS of the cloud task according to the cloud task execution time and cost that the user needs to satisfy in the process of using the resource.
Abstract
Disclosed is a genetic algorithm-based cloud computing resource scheduling method. The present invention utilizes a genetic algorithm to acquire a resource scheduling solution in cloud computing. Each of genes of a chromosome in the genetic algorithm is a subtask in a cloud task. A bit number of the chromosome indicates a number allocated to a virtual apparatus by each of the subtask representative. The present invention sets an expected Qos according to the cloud task, retains a chromosome satisfying the expected Qos in a process of population iteration and update, discarding a chromosome not satisfying the expected Qos, directly copies a chromosome having a high current population fitness value to a next generation of population according to a selection ratio, then performs crossover and mutation operations with respect to the rest of the chromosomes of the current population. The present invention ensures the quality of a chromosome in the population, enables resources in cloud computing to be utilized on-demand and flexibly expanded, arranges cloud computing resource scheduling in real time, selects an optimal policy of resource construction and adjustment, guarantees cloud computing network service quality, and improves an effectiveness ratio of the cloud computing.
Description
本发明属于云计算领域,特别涉及一种基于遗传算法的云计算资源调度方法。The invention belongs to the field of cloud computing, and particularly relates to a cloud computing resource scheduling method based on a genetic algorithm.
近年来,云计算作为一种新型的信息处理方式,已深入到我们生活工作的各个领域中。云计算是以虚拟化技术为技术支撑,最基本的理念就是按需使用,而在此过程中,由于云数据中心的能耗及其资源提供效率成为影响云计算性能的关键性问题,如何有效合理的使用云计算的资源也成了一个关键点。In recent years, cloud computing, as a new type of information processing, has penetrated into all areas of our life and work. Cloud computing is supported by virtualization technology. The most basic idea is to use it on demand. In this process, because the energy consumption of cloud data center and its resource provision efficiency become the key issues affecting cloud computing performance, how effective is it. Reasonable use of cloud computing resources has also become a key point.
随着互联网时代信息与数据的快速增长,科学、工程和商业计算领域需要处理大规模、海量的数据,对计算能力的需求远远超出自身IT架构的计算能力,这时就需要不断加大系统硬件投入来实现系统的可扩展性。另外,由于传统并行编程模型应用的局限性,客观上要求一种容易学习、使用、部署的新的并行编程框架。在这种情况下,为了节省成本和实现系统的可扩放性,云计算的概念被提了出来。云计算是分布式计算、并行处理和网格计算的进一步发展,它是基于互联网的计算,能够向各种互联网应用提供硬件服务、基础架构服务、平台服务、软件服务、存储服务的系统。通常云系统由第三方拥有的机制提供服务,用户只关心云所提供的服务。目前关于云计算系统没有统一的定义,云计算供应商根据自己企业业务推出相关的云计算战略。美国国家标准技术研究院(NIST)给出了目前权威的云计算定义:(1)云计算是一种能够通过网络以便利的、按需的方式访问一个可配置的计算资源共享池(包括网络、服务器、存储、应用和服务等)的模式,这个资源共享池能以最少的管理开销及最少的与供应商的交互,迅速配置、提供或释放资源;(2)云计算模式具有5个基本特征:按需自助服务、广泛的网络访问、共享的资源池、快速弹性能力、可度量的服务,还包括3种服务模式:软件即服务(SaaS)、平台即服务(PaaS)、基础设施即服务(IaaS),以及4种部署方式:私有云、社区云、公有云、混合云。With the rapid growth of information and data in the Internet era, the scientific, engineering, and business computing fields need to deal with large-scale, massive amounts of data. The demand for computing power far exceeds the computing power of its own IT architecture. Hardware investment to achieve system scalability. In addition, due to the limitations of traditional parallel programming model applications, a new parallel programming framework that is easy to learn, use, and deploy is objectively required. In this case, in order to save costs and achieve system scalability, the concept of cloud computing has been proposed. Cloud computing is a further development of distributed computing, parallel processing and grid computing. It is an Internet-based computing system that provides hardware services, infrastructure services, platform services, software services, and storage services to various Internet applications. Usually the cloud system is served by a mechanism owned by a third party, and the user only cares about the services provided by the cloud. At present, there is no unified definition of cloud computing systems, and cloud computing providers launch relevant cloud computing strategies based on their own business. The National Institute of Standards and Technology (NIST) gives the current definition of authoritative cloud computing: (1) Cloud computing is a convenient, on-demand access to a configurable computing resource pool (including networks) over the network. , server, storage, applications, services, etc., this resource sharing pool can quickly configure, provide or release resources with minimal administrative overhead and minimal interaction with vendors; (2) cloud computing model has 5 basics Features: On-demand self-service, extensive network access, shared resource pools, rapid resiliency, measurable services, and three service models: Software as a Service (SaaS), Platform as a Service (PaaS), Infrastructure Services (IaaS), and four deployment methods: private cloud, community cloud, public cloud, hybrid cloud.
云计算所提供的服务面向的用户群是庞大的,因此“云”中任务数量是巨大
的,系统每时每刻都出来海量的人物,所以资源管理是云计算的一个关键性问题,它的调度策略与算法直接影响着云系统的性能及成本。随着用户对云计算需求的不断增长,云数据中心规模日益庞大,其能耗问题也越来越严重。据文献报告,全球数据中心在2005年~2010年增加了约56%,美国数据中心的能耗增加了36%。一个5万个节点的数据中心每年消耗的电能超过1亿千瓦,能耗占了一个数据中心运维成本中的40%。在国内,通信运营商的数据中心是耗电大户,中国电信数据中心年耗电11.2亿千瓦时,中国联通数据中心为9.9亿千瓦时到2020年,全球主要云计算运营商的能耗将接近2万亿千瓦时。因此,研究云计算环境下的如何有效的利用资源有迫切的需求。The service provided by cloud computing is huge, so the number of tasks in the "cloud" is huge.
The system comes out with a large number of people every moment, so resource management is a key issue in cloud computing. Its scheduling strategy and algorithm directly affect the performance and cost of the cloud system. As the demand for cloud computing continues to grow, cloud data centers are becoming larger and larger, and their energy consumption problems are becoming more serious. According to the literature, the global data center increased by about 56% between 2005 and 2010, and the energy consumption of the US data center increased by 36%. A 50,000-node data center consumes more than 100 million kilowatts of electricity per year, accounting for 40% of a data center's operating costs. In China, the data center of communication operators is a large power consumer. China Telecom's data center consumes 1.12 billion kWh per year. China Unicom's data center is 990 million kWh by 2020. The energy consumption of major cloud computing operators around the world will be close. 2 trillion kWh. Therefore, there is an urgent need to study how to effectively use resources in the cloud computing environment.
发明内容Summary of the invention
本发明的目的在于克服现有技术的缺点与不足,提供一种基于遗传算法的云计算资源调度方法。该方法能够实时做出云计算资源的调度安排,决策出最佳的资源构建和调整策略,在保障云计算网络服务质量的前提下,提高了云计算的效用比。The object of the present invention is to overcome the shortcomings and shortcomings of the prior art, and to provide a cloud computing resource scheduling method based on a genetic algorithm. The method can make the scheduling of cloud computing resources in real time, determine the best resource construction and adjustment strategy, and improve the utility ratio of cloud computing under the premise of guaranteeing the service quality of the cloud computing network.
本发明的目的通过下述技术方案实现:一种基于遗传算法的云计算资源调度方法,步骤如下:The object of the present invention is achieved by the following technical solution: a cloud computing resource scheduling method based on a genetic algorithm, the steps are as follows:
S1、将用户提交的云任务切分成若干个子任务,并且根据云任务设置一个预期的Qos(Quality of Service,服务质量);预期的Qos是指用户获取云计算资源时,所花费时间和费用的一个要求;S1, the cloud task submitted by the user is divided into several subtasks, and an expected Qos (Quality of Service) is set according to the cloud task; the expected Qos refers to the time and cost of the user when acquiring the cloud computing resource. a request;
S2、初始化:初始化种群规模,并且设置种群最大迭代次数;S2, initialization: initialize the population size, and set the maximum number of iterations of the population;
S3、生成初始种群:按照编码规则随机生成染色体,将云任务抽象为染色体的编码输入,染色体中的每个基因为该云任务中的子任务,染色体的位序号表示每个子任务代表分配到虚拟机的号码;然后判断染色体是否满足预期的Qos,若满足,则将该染色体加入到初始种群中,若不满足,则丢弃该染色体,直到达到初始种群的规模;S3. Generating an initial population: randomly generating a chromosome according to an encoding rule, abstracting a cloud task into a coding input of a chromosome, each gene in the chromosome is a subtask in the cloud task, and a bit number of the chromosome indicates that each subtask representative is assigned to a virtual The number of the machine; then determine whether the chromosome satisfies the expected Qos, and if so, add the chromosome to the initial population, and if not, discard the chromosome until the size of the initial population is reached;
S4、根据适应度函数计算当前种群中每条染色体的适应度值;S4. Calculating a fitness value of each chromosome in the current population according to a fitness function;
S5、选择、交叉和变异操作:根据选择概率选择适应度值高的染色体,将这些染色体复制到下一代新种群;并且针对当前种群中剩下的染色体进行交叉和变异操作;S5, selection, crossover, and mutation operations: selecting chromosomes with high fitness values according to selection probabilities, copying these chromosomes to the next generation of new populations; and performing crossover and mutation operations on the remaining chromosomes in the current population;
S6、判断步骤S5中交叉和变异操作后的染色体是否满足预期的Qos,若满足,
则加入到下一代新种群中,若不满足,则丢弃,直到新种群达到种群规模,并并且种群迭代次数加1;S6. Determine whether the chromosome after the crossover and mutation operation in step S5 satisfies the expected QoS, if satisfied,
Then join the next generation of new populations, if not satisfied, discard until the new population reaches the population size, and the number of population iterations is increased by 1;
S7、判断种群迭代次数是否达到最大迭代次数,若否,则返回步骤S4,若是,则进入步骤S8;S7, it is determined whether the number of population iterations reaches the maximum number of iterations, if not, then returns to step S4, and if so, proceeds to step S8;
S8、将最终得到的新种群中适应度值最高的染色体作为最优解,并且针对该染色体进行解码操作得到虚拟机的号码,将该染色体作为资源调度的最优解。S8: Taking the chromosome with the highest fitness value among the newly obtained new population as the optimal solution, and performing decoding operation on the chromosome to obtain the number of the virtual machine, and using the chromosome as the optimal solution for resource scheduling.
优选的,所述步骤S3中采用基于路径的编码方式进行染色体编码。Preferably, in step S3, chromosome coding is performed by using a path-based coding method.
优选的,所述步骤S4中通过以下自适应函数计算出种群中各染色体的适应度:Preferably, in step S4, the fitness of each chromosome in the population is calculated by the following adaptive function:
f(Xk)=1/(lg(ECtotal+1));f(X k )=1/(lg(EC total +1));
其中ECtotal为染色体Xk中的资源调度方案的总能耗,f(Xk)为染色体Xk的适应度。Where EC total is the total energy consumption of the resource scheduling scheme in chromosome X k , and f(X k ) is the fitness of chromosome X k .
更进一步的,所述染色体Xk中的资源调度方案的总能耗ECtotal为:Further, the total energy consumption EC total of the resource scheduling scheme in the chromosome X k is:
其中WECj为宿主机j在一个周期内的工作能耗,mj为宿主机j上运行的子任务数;DECj为宿主机j在一个周期内休眠状态时的能耗;IECj为宿主机j空闲时的能耗;N为宿主机的个数;WEC j is the working energy consumption of the host j in one cycle, m j is the number of subtasks running on the host j; DEC j is the energy consumption of the host j in the dormant state in one cycle; IEC j is the sink The energy consumption when host j is idle; N is the number of hosts;
其中CECij为:Where CEC ij is:
CECij=CTij×PCj=CIi/CSj×PCj;CEC ij = CT ij × PC j = CI i / CS j × PC j ;
CTij表示宿主机j上子任务i的功耗,CIi表示表示宿主机j上子任务i的执行命令数,CSj表示宿主机j的CPU处理速度,PCj表示宿主机j工作时的功耗;CT ij represents the power consumption of subtask i on host j, CI i represents the number of execution commands of subtask i on host j, CS j represents the CPU processing speed of host j, and PC j represents the operation of host j Power consumption
其中SECij为:Where SEC ij is:
SECij=SDi/SSj×PSj;SEC ij =SD i /SS j ×PS j ;
SECij表示宿主机j上子任务i的存储能耗,SDi表示宿主机j上子任务i所需要读写的数据量,SSj表示宿主机j磁盘读取速率,PSj表示宿主机j存储功耗。SEC ij represents the storage energy consumption of subtask i on host j, SD i represents the amount of data that subtask i needs to read and write on host j, SS j represents the disk read rate of host j, and PS j represents host j Store power consumption.
更进一步的,所述步骤S5中的选择操作中选择概率采用轮盘算子,种群中各染色体的选择概率为:Further, the selection probability in the selection operation in the step S5 is a roulette operator, and the selection probability of each chromosome in the population is:
其中popSize为种群的规模。Among them popSize is the size of the population.
优选的,所述步骤S5中的变异操作,采用替换变异方式,先从父代染色体中选择一个子位串,然后再随机在剩下的位串中选择一个位置,并插入该子位串。Preferably, the mutation operation in the step S5 adopts a substitution variation method, first selecting a sub-bit string from the parent chromosome, and then randomly selecting a position in the remaining bit string and inserting the sub-bit string.
优选的,所述步骤S5中的变异操作采用部分匹配方法,随机选择两个交叉点,在两个点之间的位置将进行交叉,其他位置进行复制。Preferably, the mutation operation in the step S5 adopts a partial matching method, and two intersection points are randomly selected, the position between the two points will be crossed, and the other positions are copied.
优选的,所述步骤S1中设置的预期的Qos中,各宿主机中各子任务的执行时间满足以下条件:Preferably, in the expected Qos set in the step S1, the execution time of each subtask in each host satisfies the following conditions:
TEij=CTij+STij;TE ij =CT ij +ST ij ;
TEij表示宿主机j上子任务i的运行总时间,CTij表示宿主机j上子任务i的任务执行时间,STij表示宿主机j上读写数据的时间;TE ij represents the total running time of the subtask i on the host j, CT ij represents the task execution time of the subtask i on the host j, and ST ij represents the time of reading and writing data on the host j;
宿主机j中所有子任务的执行时间Tj满足以下条件:Host j in the execution time of all subtasks T j satisfies the following conditions:
mj为宿主机j中子任务的数量;m j is the number of subtasks in the host j;
云任务执行的时间Ttotal满足以下条件:The time T total of cloud task execution meets the following conditions:
N为宿主机的个数;N is the number of hosts;
所述步骤S1中设置的预期的Qos中费用满足以下条件:The expected cost in the Qos set in the step S1 satisfies the following conditions:
其中costij表示宿主机j中执行子任务i所需费用,s Re qi表示的是宿主机j子任务i对应的资源,s Pr icej表示的是宿主机j中子任务i对应的资源所需费用。Where cost ij represents the cost of executing subtask i in host j, s Re q i represents the resource corresponding to subtask i of host j, and s Pr ice j represents the resource corresponding to subtask i in host j The cost.
优选的,所述步骤S1中采用云计算中的MapReduce编程模型,将用户提交的云任务切分成若干个子任务。Preferably, in the step S1, the MapReduce programming model in the cloud computing is used to divide the cloud task submitted by the user into a plurality of subtasks.
本发明相对于现有技术具有如下的优点及效果:The present invention has the following advantages and effects over the prior art:
(1)本发明方法利用遗传算法获取到云计算中最优的资源调度方案,本发明方法中根据云任务设置预期的Qos,其中该预期的Qos为用户获取云计算资源时,所花费时间和费用的一个要求;在种群迭代更新过程中将满足预期的Qos的染色体留下,将不满足预期的Qos的染色体进行丢弃,并且根据选择概率将适应
度值高的染色体遗传到下一代种群中,保证了种群中染色体的质量,实现了云计算的资源按需使用、弹性扩展的特点,能够实时做出云计算资源的调度安排,决策出最佳的资源构建和调整策略,在保障云计算网络服务质量的前提下,提高了云计算的效用比。(1) The method of the present invention uses a genetic algorithm to obtain an optimal resource scheduling scheme in the cloud computing. In the method of the present invention, the expected QoS is set according to the cloud task, wherein the expected QoS takes time for the user to acquire the cloud computing resource. A requirement for the cost; the chromosomes that satisfy the expected Qos are left in the iterative update of the population, the chromosomes that do not satisfy the expected Qos are discarded, and will be adapted according to the selection probability
The chromosome with high degree of value is inherited into the next generation population, which ensures the quality of the chromosome in the population, realizes the on-demand use and flexible expansion of the resources of cloud computing, and can make the scheduling of cloud computing resources in real time, making the best decision. The resource construction and adjustment strategy improves the utility ratio of cloud computing on the premise of guaranteeing the quality of cloud computing network services.
(2)本发明方法中通过与染色体中的资源调度方案总能耗相关的自适应函数计算染色体的适应度,该适应度函数能够评价染色体的优劣状态,其适应度值越大,表示染色体的适应度越好,反之越差,通过该适应度函数能够更加准确的选择出符合实际最优资源调度要求的染色体,进一步保证了遗传算法最终得到的资源调度方案为最优解。(2) In the method of the present invention, the fitness of the chromosome is calculated by an adaptive function related to the total energy consumption of the resource scheduling scheme in the chromosome, and the fitness function can evaluate the pros and cons of the chromosome, and the larger the fitness value, the chromosome The better the fitness, and vice versa, the more accurate the chromosomes that meet the actual optimal resource scheduling requirements can be selected by the fitness function, which further ensures that the resource scheduling scheme finally obtained by the genetic algorithm is the optimal solution.
图1是本发明方法流程图。Figure 1 is a flow chart of the method of the present invention.
下面结合实施例及附图对本发明作进一步详细的描述,但本发明的实施方式不限于此。The present invention will be further described in detail below with reference to the embodiments and drawings, but the embodiments of the present invention are not limited thereto.
实施例Example
如图1所示,本实施例公开了一种基于遗传算法的云计算资源调度方法,步骤如下:As shown in FIG. 1 , this embodiment discloses a cloud computing resource scheduling method based on a genetic algorithm, and the steps are as follows:
S1、采用云计算的MapReduce编程模型,将用户提交的云任务切分成若干个子任务,并且根据云任务设置一个预期的Qos;预期的Qos是指用户获取云计算资源时,所花费时间和费用的一个要求。S1: Using the MapReduce programming model of cloud computing, the cloud task submitted by the user is divided into several subtasks, and an expected Qos is set according to the cloud task; the expected Qos refers to the time and cost of the user when acquiring the cloud computing resource. A request.
S2、初始化:初始化种群规模popSize,并且设置种群最大迭代次数Tmax。S2, initialization: Initialize the population size popSize, and set the population maximum iteration number Tmax.
S3、生成初始种群:按照编码规则随机生成染色体,将云任务抽象为染色体的编码输入,染色体中的每个基因为该云任务中的子任务,染色体的位序号表示每个子任务分配到虚拟机的号码;然后判断染色体是否满足预期的Qos,若满足,则将该染色体加入到初始种群中,若不满足,则丢弃该染色体,直到达到初始种群的规模。在本实施例中采用了基于路径的编码方式进行染色体的编码,当然也可以采用基于二进制、基于矩阵、基于邻接、基于索引等的编码方式。S3. Generating an initial population: randomly generating a chromosome according to an encoding rule, abstracting the cloud task into a coding input of the chromosome, each gene in the chromosome is a subtask in the cloud task, and the bit number of the chromosome indicates that each subtask is assigned to the virtual machine. The number is then determined whether the chromosome satisfies the expected QoS, and if so, the chromosome is added to the initial population, and if not, the chromosome is discarded until the size of the initial population is reached. In the present embodiment, the path-based coding method is used to perform coding of a chromosome, and of course, a coding method based on binary, matrix-based, contiguous, index-based, etc. may be employed.
S4、根据适应度函数计算当前种群中每条染色体的适应度;其中本步骤中
通过以下自适应函数计算出种群中各染色体的适应度f(Xk):S4. Calculating the fitness of each chromosome in the current population according to the fitness function; wherein in this step, the fitness f(X k ) of each chromosome in the population is calculated by the following adaptive function:
f(Xk)=1/(lg(ECtotal+1));f(X k )=1/(lg(EC total +1));
其中ECtotal为染色体Xk中的资源调度方案的总能耗;适应度函数评价了染色体的优劣,其适应度值越大,表示染色体的适应度越好,反之越差。在本步骤中ECtotal越小,适应度值越大。EC total is the total energy consumption of the resource scheduling scheme in chromosome X k ; the fitness function evaluates the pros and cons of the chromosome, and the greater the fitness value, the better the fitness of the chromosome, and vice versa. The smaller the EC total in this step, the larger the fitness value.
染色体Xk中的资源调度方案的总能耗ECtotal为:The total energy consumption EC total of the resource scheduling scheme in chromosome X k is:
其中WECj为宿主机j在一个周期内的工作能耗,mj为宿主机j上运行的子任务数;DECj为宿主机j在一个周期内休眠状态时的能耗;IECj为宿主机j空闲时的能耗;N为宿主机的个数;WEC j is the working energy consumption of the host j in one cycle, m j is the number of subtasks running on the host j; DEC j is the energy consumption of the host j in the dormant state in one cycle; IEC j is the sink The energy consumption when host j is idle; N is the number of hosts;
其中CECij为:Where CEC ij is:
CECij=CTij×PCj=CIi/CSj×PCj;CEC ij = CT ij × PC j = CI i / CS j × PC j ;
CTij表示宿主机j上子任务i的功耗,CIi表示表示宿主机j上子任务i的执行命令数,CSj表示宿主机j的CPU处理速度,PCj表示宿主机j工作时的功耗;CT ij represents the power consumption of subtask i on host j, CI i represents the number of execution commands of subtask i on host j, CS j represents the CPU processing speed of host j, and PC j represents the operation of host j Power consumption
其中SECij为:Where SEC ij is:
SECij=SDi/SSj×PSj;SEC ij =SD i /SS j ×PS j ;
SECij表示宿主机j上子任务i的存储能耗,SDi表示宿主机j上子任务i所需要读写的数据量,SSj表示宿主机j磁盘读取速率,PSj表示宿主机j存储功耗。SEC ij represents the storage energy consumption of subtask i on host j, SD i represents the amount of data that subtask i needs to read and write on host j, SS j represents the disk read rate of host j, and PS j represents host j Store power consumption.
S5、选择、交叉和变异操作:根据选择概率选择适应度值高的染色体,将这些适应度高的染色体复制到下一代新种群;并且针对当前种群中剩下的染色体进行交叉和变异操作。S5, selection, crossover and mutation operations: selecting chromosomes with high fitness values according to the selection probability, copying these highly adaptive chromosomes to the next generation of new populations; and performing crossover and mutation operations on the remaining chromosomes in the current population.
在本步骤中选择操作中选择概率采用轮盘算子,种群中染色体被选中的概率与其适应度成正比;种群中各染色体的选择概率为:In this step, the selection probability in the selection operation uses the roulette operator, and the probability that the chromosome in the population is selected is proportional to its fitness; the selection probability of each chromosome in the population is:
其中popSize为种群的规模,f(Xk)为染色体Xk的适应度。Where popSize is the size of the population and f(X k ) is the fitness of the chromosome X k .
本步骤中变异操作,采用替换变异方式,先从父代染色体中选择一个子位串,然后再随机在剩下的位串中选择一个位置,并插入该子位串。当然本实施例中变异操作也可以采用交换变异(EM)、插入变异(IM)、简单倒位变异(SIM),
倒位变异(IVM)、争夺变异(SM)等变异方式。In the mutation operation in this step, the substitution variation method is adopted, first selecting a sub-bit string from the parent chromosome, and then randomly selecting a position in the remaining bit string and inserting the sub-bit string. Of course, the mutation operation in this embodiment may also adopt exchange mutation (EM), insertion mutation (IM), and simple inversion variation (SIM).
Variations such as inversion variation (IVM) and competition for variation (SM).
本步骤中变异操作采用部分匹配方法,随机选择两个交叉点,在两个点之间的位置将进行交叉,其他位置进行复制。当然本实施例中交叉操作也可以采用部分匹配方法(PMX)、循环交叉法(CX)、次序交叉法(OX)、基于位置的交叉法(POS)等。In this step, the mutation operation adopts a partial matching method, and two intersection points are randomly selected, and the position between the two points will be crossed, and the other positions are copied. Of course, the cross operation in this embodiment may also adopt a partial matching method (PMX), a cyclic cross method (CX), an order cross method (OX), a position based cross method (POS), and the like.
S6、判断步骤S5中交叉和变异操作后的染色体是否满足预期的Qos,若满足,则加入到下一代新种群中,若不满足,则丢弃,直到新种群达到种群规模popSize,并且种群迭代次数T加1。S6. Determine whether the chromosome after the crossover and mutation operation in step S5 satisfies the expected QoS, and if yes, join the next generation new population, and if not, discard until the new population reaches the population size popSize, and the number of population iterations T plus 1.
S7、判断种群迭代次数是否达到最大迭代次数Tmax,若否,则返回步骤S4,若是,则进入步骤S8。S7. Determine whether the number of population iterations reaches the maximum number of iterations Tmax. If not, return to step S4, and if yes, proceed to step S8.
S8、将最终得到的新种群中适应度值最高的染色体作为最优解,并且针对该染色体进行解码操作得到虚拟机的号码,将该染色体作为资源调度的最优解。S8: Taking the chromosome with the highest fitness value among the newly obtained new population as the optimal solution, and performing decoding operation on the chromosome to obtain the number of the virtual machine, and using the chromosome as the optimal solution for resource scheduling.
云计算的Qos是指,消费者获取云计算资源时,希望时间,成本等指标能够满足消费者的预期目标。当消费者依靠资源提供者来满足其计算需求时,为了匹配他们的要求,用户任务多维Qos需求(时间,成本)必须得到保证。Cloud computing Qos means that when consumers get cloud computing resources, they hope that time, cost and other indicators can meet the expected goals of consumers. When consumers rely on resource providers to meet their computing needs, in order to match their requirements, the multi-dimensional Qos requirements (time, cost) of the user task must be guaranteed.
物力资源的能耗等于功耗与时间的乘积。资源在不同情况下的功耗也是不同的,物理主机通常分为休眠,空闲和工作三个状态。本实施例上述步骤S1中设置的预期的Qos中,各宿主机中各子任务的执行时间满足以下条件:The energy consumption of material resources is equal to the product of power consumption and time. The power consumption of resources in different situations is also different. Physical hosts are usually divided into three states: sleep, idle, and work. In the expected Qos set in the above step S1 in this embodiment, the execution time of each subtask in each host satisfies the following conditions:
TEij=CTij+STij;TE ij =CT ij +ST ij ;
TEij表示宿主机j上子任务i的运行总时间,CTij表示宿主机j上子任务i的任务执行时间,STij表示宿主机j上读写数据的时间;TE ij represents the total running time of the subtask i on the host j, CT ij represents the task execution time of the subtask i on the host j, and ST ij represents the time of reading and writing data on the host j;
宿主机j中所有子任务的执行时间Tj满足以下条件:Host j in the execution time of all subtasks T j satisfies the following conditions:
mj为宿主机j中子任务的数量;m j is the number of subtasks in the host j;
云任务执行的时间Ttotal满足以下条件:The time T total of cloud task execution meets the following conditions:
N为宿主机的个数;N is the number of hosts;
所述步骤S1中设置的预期的Qos中费用满足以下条件:The expected cost in the Qos set in the step S1 satisfies the following conditions:
其中costij表示宿主机j中执行子任务i所需费用,s Re qi表示的是宿主机j子任务i对应的资源,s Pr icej表示的是宿主机j中子任务i对应的资源所需费用。Where cost ij represents the cost of executing subtask i in host j, s Re q i represents the resource corresponding to subtask i of host j, and s Pr ice j represents the resource corresponding to subtask i in host j The cost.
资源调度方案就是时间和费用的一个权衡,本实施例根据用户在使用资源过程中上述需要满足的云任务执行时间和费用得出一个预期的时间和费用作为云任务的Qos。The resource scheduling solution is a trade-off between time and cost. In this embodiment, an expected time and cost is obtained as the QoS of the cloud task according to the cloud task execution time and cost that the user needs to satisfy in the process of using the resource.
上述实施例为本发明较佳的实施方式,但本发明的实施方式并不受上述实施例的限制,其他的任何未背离本发明的精神实质与原理下所作的改变、修饰、替代、组合、简化,均应为等效的置换方式,都包含在本发明的保护范围之内。
The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and combinations thereof may be made without departing from the spirit and scope of the invention. Simplifications should all be equivalent replacements and are included in the scope of the present invention.
Claims (9)
- 一种基于遗传算法的云计算资源调度方法,其特征在于,步骤如下:A cloud computing resource scheduling method based on genetic algorithm, characterized in that the steps are as follows:S1、将用户提交的云任务切分成若干个子任务,并且根据云任务设置一个预期的Qos;预期的Qos是指用户获取云计算资源时,所花费时间和费用的一个要求;S1: The cloud task submitted by the user is divided into a plurality of subtasks, and an expected Qos is set according to the cloud task; the expected Qos is a requirement for the time and cost of the user when acquiring the cloud computing resource;S2、初始化:初始化种群规模,并且设置种群最大迭代次数;S2, initialization: initialize the population size, and set the maximum number of iterations of the population;S3、生成初始种群:按照编码规则随机生成染色体,将云任务抽象为染色体的编码输入,染色体中的每个基因为该云任务中的子任务,染色体的位序号表示每个子任务代表分配到虚拟机的号码;然后判断染色体是否满足预期的Qos,若满足,则将该染色体加入到初始种群中,若不满足,则丢弃该染色体,直到达到初始种群的规模;S3. Generating an initial population: randomly generating a chromosome according to an encoding rule, abstracting a cloud task into a coding input of a chromosome, each gene in the chromosome is a subtask in the cloud task, and a bit number of the chromosome indicates that each subtask representative is assigned to a virtual The number of the machine; then determine whether the chromosome satisfies the expected Qos, and if so, add the chromosome to the initial population, and if not, discard the chromosome until the size of the initial population is reached;S4、根据适应度函数计算当前种群中每条染色体的适应度值;S4. Calculating a fitness value of each chromosome in the current population according to a fitness function;S5、选择、交叉和变异操作:根据选择概率选择适应度值高的染色体,将这些染色体复制到下一代新种群;并且针对当前种群中剩下的染色体进行交叉和变异操作;S5, selection, crossover, and mutation operations: selecting chromosomes with high fitness values according to selection probabilities, copying these chromosomes to the next generation of new populations; and performing crossover and mutation operations on the remaining chromosomes in the current population;S6、判断步骤S5中交叉和变异操作后的染色体是否满足预期的Qos,若满足,则加入到下一代新种群中,若不满足,则丢弃,直到新种群达到种群规模,并并且种群迭代次数加1;S6. Determine whether the chromosome after the crossover and mutation operation in step S5 satisfies the expected QoS, and if yes, join the next generation new population, and if not, discard until the new population reaches the population size, and the number of population iterations plus 1;S7、判断种群迭代次数是否达到最大迭代次数,若否,则返回步骤S4,若是,则进入步骤S8;S7, it is determined whether the number of population iterations reaches the maximum number of iterations, if not, then returns to step S4, and if so, proceeds to step S8;S8、将最终得到的新种群中适应度值最高的染色体作为最优解,并且针对该染色体进行解码操作得到虚拟机的号码,将该染色体作为资源调度的最优解。S8: Taking the chromosome with the highest fitness value among the newly obtained new population as the optimal solution, and performing decoding operation on the chromosome to obtain the number of the virtual machine, and using the chromosome as the optimal solution for resource scheduling.
- 根据权利要求1所述的基于遗传算法的云计算资源调度方法,其特征在于,所述步骤S3中采用基于路径的编码方式进行染色体编码。The genetic algorithm-based cloud computing resource scheduling method according to claim 1, wherein the step S3 uses a path-based coding method for chromosome coding.
- 根据权利要求1所述的基于遗传算法的云计算资源调度方法,其特征在于,所述步骤S4中通过以下自适应函数计算出种群中各染色体的适应度:The genetic algorithm-based cloud computing resource scheduling method according to claim 1, wherein in step S4, the fitness of each chromosome in the population is calculated by the following adaptive function:f(Xk)=1/(lg(ECtotal+1));f(X k )=1/(lg(EC total +1));其中ECtotal为染色体Xk中的资源调度方案的总能耗,f(Xk)为染色体Xk的适应度。Where EC total is the total energy consumption of the resource scheduling scheme in chromosome X k , and f(X k ) is the fitness of chromosome X k .
- 根据权利要求3所述的基于遗传算法的云计算资源调度方法,其特征在 于,所述染色体Xk中的资源调度方案的总能耗ECtotal为:The genetic algorithm-based cloud computing resource scheduling method according to claim 3, wherein the total energy consumption EC total of the resource scheduling scheme in the chromosome X k is:其中WECj为宿主机j在一个周期内的工作能耗,mj为宿主机j上运行的子任务数;DECj为宿主机j在一个周期内休眠状态时的能耗;IECj为宿主机j空闲时的能耗;N为宿主机的个数;WEC j is the working energy consumption of the host j in one cycle, m j is the number of subtasks running on the host j; DEC j is the energy consumption of the host j in the dormant state in one cycle; IEC j is the sink The energy consumption when host j is idle; N is the number of hosts;其中CECij为:Where CEC ij is:CECij=CTij×PCj=CIi/CSj×PCj;CEC ij = CT ij × PC j = CI i / CS j × PC j ;CTij表示宿主机j上子任务i的功耗,CIi表示表示宿主机j上子任务i的执行命令数,CSj表示宿主机j的CPU处理速度,PCj表示宿主机j工作时的功耗;CT ij represents the power consumption of subtask i on host j, CI i represents the number of execution commands of subtask i on host j, CS j represents the CPU processing speed of host j, and PC j represents the operation of host j Power consumption其中SECij为:Where SEC ij is:SECij=SDi/SSj×PSj;SEC ij =SD i /SS j ×PS j ;SECij表示宿主机j上子任务i的存储能耗,SDi表示宿主机j上子任务i所需要读写的数据量,SSj表示宿主机j磁盘读取速率,PSj表示宿主机j存储功耗。SEC ij represents the storage energy consumption of subtask i on host j, SD i represents the amount of data that subtask i needs to read and write on host j, SS j represents the disk read rate of host j, and PS j represents host j Store power consumption.
- 根据权利要求3所述的基于遗传算法的云计算资源调度方法,其特征在于,所述步骤S5中的选择操作中选择概率采用轮盘算子,种群中各染色体的选择概率为:The method for scheduling a cloud computing resource based on a genetic algorithm according to claim 3, wherein the selection probability in the selecting operation in the step S5 is a roulette operator, and the selection probability of each chromosome in the population is:其中popSize为种群的规模。Among them popSize is the size of the population.
- 根据权利要求1所述的基于遗传算法的云计算资源调度方法,其特征在于,所述步骤S5中的变异操作,采用替换变异方式,先从父代染色体中选择一个子位串,然后再随机在剩下的位串中选择一个位置,并插入该子位串。The genetic algorithm-based cloud computing resource scheduling method according to claim 1, wherein the mutation operation in the step S5 adopts a substitution variation method, first selecting a sub-bit string from the parent chromosome, and then randomly Select a position in the remaining bit string and insert the sub-bit string.
- 根据权利要求1所述的基于遗传算法的云计算资源调度方法,其特征在于,所述步骤S5中的变异操作采用部分匹配方法,随机选择两个交叉点,在两个点之间的位置将进行交叉,其他位置进行复制。The genetic algorithm-based cloud computing resource scheduling method according to claim 1, wherein the mutation operation in the step S5 adopts a partial matching method, and two intersection points are randomly selected, and the position between the two points will be Cross and other locations to copy.
- 根据权利要求1所述的基于遗传算法的云计算资源调度方法,其特征在于,所述步骤S1中设置的预期的Qos中,各宿主机中各子任务的执行时间满足以下条件:The method for scheduling a cloud computing resource based on a genetic algorithm according to claim 1, wherein in the expected Qos set in step S1, the execution time of each subtask in each host satisfies the following conditions:TEij=CTij+STij;TE ij =CT ij +ST ij ;TEij表示宿主机j上子任务i的运行总时间,CTij表示宿主机j上子任务i的任 务执行时间,STij表示宿主机j上读写数据的时间;TE ij represents the total running time of the subtask i on the host j, CT ij represents the task execution time of the subtask i on the host j, and ST ij represents the time of reading and writing data on the host j;宿主机j中所有子任务的执行时间Tj满足以下条件:Host j in the execution time of all subtasks T j satisfies the following conditions:mj为宿主机j中子任务的数量;m j is the number of subtasks in the host j;云任务执行的时间Ttotal满足以下条件:The time T total of cloud task execution meets the following conditions:N为宿主机的个数;N is the number of hosts;所述步骤S1中设置的预期的Qos中费用满足以下条件:The expected cost in the Qos set in the step S1 satisfies the following conditions:其中costij表示宿主机j中执行子任务i所需费用,sReqi表示的是宿主机j子任务i对应的资源,sPricej表示的是宿主机j中子任务i对应的资源所需费用。Where cost ij represents the cost of executing subtask i in host j, sReq i represents the resource corresponding to subtask i of host j, and sPrice j represents the cost of the resource corresponding to subtask i in host j.
- 根据权利要求1所述的基于遗传算法的云计算资源调度方法,其特征在于,所述步骤S1中采用云计算中的MapReduce编程模型,将用户提交的云任务切分成若干个子任务。 The genetic algorithm-based cloud computing resource scheduling method according to claim 1, wherein the step S1 uses a MapReduce programming model in the cloud computing to divide the cloud task submitted by the user into a plurality of subtasks.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510185184.8 | 2015-04-17 | ||
CN201510185184.8A CN104811491A (en) | 2015-04-17 | 2015-04-17 | Cloud computing resource scheduling method based on genetic algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2016165392A1 true WO2016165392A1 (en) | 2016-10-20 |
Family
ID=53695979
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2015/098620 WO2016165392A1 (en) | 2015-04-17 | 2015-12-24 | Genetic algorithm-based cloud computing resource scheduling method |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN104811491A (en) |
WO (1) | WO2016165392A1 (en) |
Cited By (57)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107995039A (en) * | 2017-12-07 | 2018-05-04 | 福州大学 | The resource self study of facing cloud software service and self-adapting distribution method |
CN108762927A (en) * | 2018-05-29 | 2018-11-06 | 武汉轻工大学 | The multiple target method for scheduling task of mobile cloud computing |
CN108958919A (en) * | 2018-07-13 | 2018-12-07 | 湘潭大学 | More DAG task schedule expense fairness assessment models of limited constraint in a kind of cloud computing |
CN109344953A (en) * | 2018-08-23 | 2019-02-15 | 东华理工大学 | Cloud service combined method |
CN109343966A (en) * | 2018-11-01 | 2019-02-15 | 西北工业大学 | A kind of cluster organization method and device of unmanned node |
CN109376396A (en) * | 2018-09-27 | 2019-02-22 | 中国科学技术大学 | Road network generally investigates facility addressing points distributing method |
CN109542623A (en) * | 2018-11-22 | 2019-03-29 | 中国人民解放军陆军工程大学 | The Optimization deployment method of virtual system in the server |
CN109901932A (en) * | 2019-03-12 | 2019-06-18 | 东北大学 | A kind of Server Consolidation method based on virtual machine |
CN110008023A (en) * | 2019-03-26 | 2019-07-12 | 湖南农业大学 | Cloud computing system budgetary restraints Random Task dispatching method based on genetic algorithm |
CN110321217A (en) * | 2019-03-25 | 2019-10-11 | 深圳大学 | A kind of cloud resource dispatching method, device, equipment and the storage medium of multiple target |
CN110363298A (en) * | 2019-07-19 | 2019-10-22 | 中建安装集团有限公司 | Intelligent adjustment of route alignment and gradient method based on evolution algorithm |
CN110569589A (en) * | 2019-08-30 | 2019-12-13 | 西南交通大学 | Aisle arrangement method |
CN110851257A (en) * | 2019-11-07 | 2020-02-28 | 中国石油大学(华东) | Genetic and differential hybrid evolution cloud computing task scheduling algorithm based on early-stage catastrophe strategy |
CN110851277A (en) * | 2019-11-08 | 2020-02-28 | 中国石油大学(华东) | Task scheduling strategy based on edge cloud cooperation in augmented reality scene |
CN110942811A (en) * | 2019-12-14 | 2020-03-31 | 华东理工大学 | Design method of catalyst in surface catalytic reaction based on genetic algorithm |
CN111079987A (en) * | 2019-11-28 | 2020-04-28 | 电子科技大学 | Semiconductor workshop production scheduling method based on genetic algorithm |
CN111126707A (en) * | 2019-12-26 | 2020-05-08 | 华自科技股份有限公司 | Energy consumption equation construction and energy consumption prediction method and device |
CN111291888A (en) * | 2020-01-21 | 2020-06-16 | 西安科技大学 | Scheduling optimization method for airport special vehicles |
CN111325648A (en) * | 2020-02-20 | 2020-06-23 | 杭州电子科技大学 | Aging service combination optimization method and system based on hesitation ambiguity |
CN111325481A (en) * | 2020-03-16 | 2020-06-23 | 郑州大学 | Intelligent automatic container terminal energy-saving comprehensive scheduling method |
CN111611748A (en) * | 2020-05-25 | 2020-09-01 | 上海大学 | Data-driven material reverse design method and system |
CN111626516A (en) * | 2020-05-30 | 2020-09-04 | 湖南科技大学 | Double-deep-position four-way shuttle system order ordering optimization method considering goods reversing strategy |
CN111651434A (en) * | 2019-10-22 | 2020-09-11 | 摩登汽车有限公司 | Millimeter wave radar data processing method, device and equipment based on genetic algorithm |
CN111652373A (en) * | 2020-05-29 | 2020-09-11 | 济南浪潮高新科技投资发展有限公司 | Task scheduling method, device, equipment and storage medium of cloud robot |
CN111679907A (en) * | 2020-05-14 | 2020-09-18 | 西安交通大学 | Cloud computing task scheduling method |
CN111724044A (en) * | 2020-05-29 | 2020-09-29 | 浙江汉脑数码科技有限公司 | Intelligent production scheduling system based on genetic algorithm |
CN111737120A (en) * | 2020-06-15 | 2020-10-02 | 河南工业大学 | Software defect repairing method and device |
CN111985647A (en) * | 2020-07-21 | 2020-11-24 | 西安理工大学 | Printing bookbinding job scheduling optimization method based on genetic algorithm |
CN112148443A (en) * | 2020-09-02 | 2020-12-29 | 安徽中科龙安科技股份有限公司 | Task optimization scheduling method and system of Loongson big data all-in-one machine |
CN112181655A (en) * | 2020-09-30 | 2021-01-05 | 杭州电子科技大学 | Hybrid genetic algorithm-based calculation unloading method in mobile edge calculation |
CN112214320A (en) * | 2020-10-10 | 2021-01-12 | 福州大学 | PSO-GA-based cloud service resource self-adaptive distribution method facing load time window |
CN112231117A (en) * | 2020-10-13 | 2021-01-15 | 山东大学 | Cloud robot service selection method and system based on dynamic vector hybrid genetic algorithm |
CN112256402A (en) * | 2020-10-30 | 2021-01-22 | 深圳供电局有限公司 | Cloud platform data center resource prediction and scheduling method and system |
CN112272381A (en) * | 2020-10-22 | 2021-01-26 | 大连大学 | Satellite network task deployment method and system |
CN112272102A (en) * | 2020-09-11 | 2021-01-26 | 北京工业大学 | Method and device for unloading and scheduling edge network service |
CN112486651A (en) * | 2020-11-30 | 2021-03-12 | 中国电子科技集团公司第十五研究所 | Cloud test platform task scheduling method based on improved genetic algorithm |
CN112578436A (en) * | 2019-09-27 | 2021-03-30 | 中国石油化工股份有限公司 | Wavelet extraction method and system |
CN112651538A (en) * | 2019-10-11 | 2021-04-13 | 阿里巴巴集团控股有限公司 | Method and device for configuring rack |
CN112731378A (en) * | 2020-12-15 | 2021-04-30 | 郑州大学 | Method for acquiring dielectric constant by ant colony genetic combination, radar detection method and system |
CN112749776A (en) * | 2021-01-12 | 2021-05-04 | 南京信息工程大学 | Job shop scheduling method based on improved hybrid genetic algorithm |
CN112787833A (en) * | 2019-11-07 | 2021-05-11 | 中国电信股份有限公司 | Method and device for deploying CDN (content delivery network) server |
CN112884368A (en) * | 2021-03-23 | 2021-06-01 | 合肥工业大学 | Multi-target scheduling method and system for minimizing delivery time and delay of high-end equipment |
CN112949195A (en) * | 2021-03-10 | 2021-06-11 | 国网吉林省电力有限公司信息通信公司 | Multi-target task scheduling method for power cloud data center |
CN113132445A (en) * | 2020-01-10 | 2021-07-16 | 阿里巴巴集团控股有限公司 | Resource scheduling method, device, network system and storage medium |
CN113487142A (en) * | 2021-06-15 | 2021-10-08 | 昆山翦统智能科技有限公司 | Evolution optimization method and system for E-government performance assessment management |
CN113641471A (en) * | 2021-07-30 | 2021-11-12 | 平安科技(深圳)有限公司 | Soft load scheduling method, device, equipment and medium based on genetic algorithm model |
CN114189914A (en) * | 2021-11-04 | 2022-03-15 | 国网江苏省电力有限公司南京供电分公司 | WiFi6 optimized switching decision method and system based on service priority |
CN114285880A (en) * | 2021-12-27 | 2022-04-05 | 电子科技大学长三角研究院(衢州) | Internet of things lightweight equipment intelligentization method based on enhanced twinning |
CN114443258A (en) * | 2020-10-30 | 2022-05-06 | 中国移动通信集团浙江有限公司 | Resource scheduling method, device and equipment for virtual machine and storage medium |
CN114528088A (en) * | 2022-04-25 | 2022-05-24 | 北京航天驭星科技有限公司 | Satellite detection task processing method and device, electronic equipment and storage medium |
CN114840326A (en) * | 2022-06-01 | 2022-08-02 | 中国电信股份有限公司 | Resource scheduling method and device, electronic equipment and storage medium |
CN115185655A (en) * | 2022-06-23 | 2022-10-14 | 郑州轻工业大学 | Genetic task scheduling method based on gene frequency improvement |
CN115733839A (en) * | 2022-11-14 | 2023-03-03 | 中国电子科技集团公司第五十四研究所 | Bandwidth self-adaptive data chain intelligent distribution strategy optimization method |
CN116307296A (en) * | 2023-05-22 | 2023-06-23 | 南京航空航天大学 | Cloud resource optimization configuration method |
CN116922402A (en) * | 2023-09-19 | 2023-10-24 | 杭州芯控智能科技有限公司 | Genetic algorithm-based multi-mechanical arm multi-task distribution method, system and medium |
CN116968037A (en) * | 2023-09-21 | 2023-10-31 | 杭州芯控智能科技有限公司 | Multi-mechanical-arm cooperative task scheduling method |
CN117290077A (en) * | 2023-11-24 | 2023-12-26 | 深圳市飞梵实业有限公司 | Medical platform cloud computing task scheduling method and system based on genetic algorithm |
Families Citing this family (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104811491A (en) * | 2015-04-17 | 2015-07-29 | 华南理工大学 | Cloud computing resource scheduling method based on genetic algorithm |
CN105740051B (en) * | 2016-01-27 | 2019-03-22 | 北京工业大学 | Cloud computing resources based on Revised genetic algorithum dispatch implementation method |
CN106502908A (en) * | 2016-11-07 | 2017-03-15 | 牡丹江师范学院 | Executable basic path evolution generation method based on statistical analysis |
CN107093002A (en) * | 2017-03-02 | 2017-08-25 | 平顶山天安煤业股份有限公司 | A kind of bore closed quality classification and hazard assessment system based on cloud computing |
CN107528799A (en) * | 2017-08-23 | 2017-12-29 | 湖北工业大学 | A kind of cloud computing resources distribution method and system |
CN108897600A (en) * | 2018-06-14 | 2018-11-27 | 郑州云海信息技术有限公司 | A kind of virtual machine placement method under cloud computing environment |
CN108881432A (en) * | 2018-06-15 | 2018-11-23 | 广东省城乡规划设计研究院 | Cloud computing cluster load dispatching method based on GA algorithm |
CN109343933B (en) * | 2018-09-17 | 2021-11-23 | 浙江工业大学 | Virtual machine initial placement strategy method based on improved genetic algorithm |
CN109615074B (en) * | 2018-12-14 | 2023-02-17 | 北京字节跳动网络技术有限公司 | Method and device for generating monster configuration of game |
CN111832725B (en) * | 2019-04-15 | 2023-05-26 | 电子科技大学 | Multi-robot multi-task allocation method and device based on improved genetic algorithm |
CN110321208B (en) * | 2019-06-27 | 2022-10-14 | 大连理工大学 | Evolutionary computing method for solving cloud task scheduling |
CN110413231A (en) * | 2019-07-16 | 2019-11-05 | 东华大学 | The data-optimized storage method of inexpensive high availability under a kind of cloudy environment |
CN112990515A (en) * | 2019-12-02 | 2021-06-18 | 中船重工信息科技有限公司 | Workshop resource scheduling method based on heuristic optimization algorithm |
CN111324422B (en) * | 2020-02-24 | 2024-04-16 | 武汉轻工大学 | Multi-target virtual machine deployment method, device, equipment and storage medium |
CN111381936B (en) * | 2020-03-23 | 2023-03-31 | 中山大学 | Method and system for allocating service container resources under distributed cloud system-cloud cluster architecture |
CN111800310B (en) * | 2020-06-17 | 2021-11-26 | 北京智网信测科技有限公司 | Scheduling method for task management module of Internet of things test cloud platform |
CN111984403A (en) * | 2020-08-05 | 2020-11-24 | 华东计算技术研究所(中国电子科技集团公司第三十二研究所) | Method and system for distributing and deploying software components in distributed system |
CN112256422B (en) * | 2020-11-17 | 2023-08-04 | 中国人民解放军战略支援部队信息工程大学 | Heterogeneous platform task scheduling method and system based on Q learning |
CN115412930A (en) * | 2021-05-26 | 2022-11-29 | 维沃移动通信有限公司 | Policy generation method, device, terminal, equipment and session management unit |
CN113515351B (en) * | 2021-09-07 | 2022-03-29 | 华南理工大学 | Resource scheduling implementation method based on energy consumption and QoS (quality of service) cooperative optimization |
CN114214869B (en) * | 2021-12-17 | 2023-08-15 | 浙江华章科技有限公司 | Sizing optimization method of film transfer sizing machine |
CN114518937B (en) * | 2022-01-27 | 2022-11-04 | 广州鼎甲计算机科技有限公司 | Virtual machine backup scheduling method, system, device and storage medium |
CN114745310B (en) * | 2022-03-31 | 2024-01-12 | 工银科技有限公司 | Flow threshold determining method and device based on genetic algorithm |
CN115421885B (en) * | 2022-08-29 | 2023-04-28 | 山东财经大学 | Distributed multi-target cloud task scheduling method and device and cloud service system |
CN117130790B (en) * | 2023-10-23 | 2023-12-29 | 云南蓝队云计算有限公司 | Dynamic scheduling method for cloud computing resource pool |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103024048A (en) * | 2012-12-17 | 2013-04-03 | 南京邮电大学 | Resources scheduling method under cloud environment |
CN103902375A (en) * | 2014-04-11 | 2014-07-02 | 北京工业大学 | Cloud task scheduling method based on improved genetic algorithm |
CN104811491A (en) * | 2015-04-17 | 2015-07-29 | 华南理工大学 | Cloud computing resource scheduling method based on genetic algorithm |
-
2015
- 2015-04-17 CN CN201510185184.8A patent/CN104811491A/en active Pending
- 2015-12-24 WO PCT/CN2015/098620 patent/WO2016165392A1/en active Application Filing
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103024048A (en) * | 2012-12-17 | 2013-04-03 | 南京邮电大学 | Resources scheduling method under cloud environment |
CN103902375A (en) * | 2014-04-11 | 2014-07-02 | 北京工业大学 | Cloud task scheduling method based on improved genetic algorithm |
CN104811491A (en) * | 2015-04-17 | 2015-07-29 | 华南理工大学 | Cloud computing resource scheduling method based on genetic algorithm |
Cited By (89)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107995039B (en) * | 2017-12-07 | 2020-11-03 | 福州大学 | Resource self-learning and self-adaptive distribution method for cloud software service |
CN107995039A (en) * | 2017-12-07 | 2018-05-04 | 福州大学 | The resource self study of facing cloud software service and self-adapting distribution method |
CN108762927A (en) * | 2018-05-29 | 2018-11-06 | 武汉轻工大学 | The multiple target method for scheduling task of mobile cloud computing |
CN108762927B (en) * | 2018-05-29 | 2022-01-14 | 上海艾涛信息科技发展有限公司 | Multi-target task scheduling method for mobile cloud computing |
CN108958919A (en) * | 2018-07-13 | 2018-12-07 | 湘潭大学 | More DAG task schedule expense fairness assessment models of limited constraint in a kind of cloud computing |
CN109344953A (en) * | 2018-08-23 | 2019-02-15 | 东华理工大学 | Cloud service combined method |
CN109376396A (en) * | 2018-09-27 | 2019-02-22 | 中国科学技术大学 | Road network generally investigates facility addressing points distributing method |
CN109376396B (en) * | 2018-09-27 | 2022-10-28 | 中国科学技术大学 | Site selection and point distribution method for road network general survey facilities |
CN109343966A (en) * | 2018-11-01 | 2019-02-15 | 西北工业大学 | A kind of cluster organization method and device of unmanned node |
CN109343966B (en) * | 2018-11-01 | 2023-04-07 | 西北工业大学 | Cluster organization method and device for unmanned nodes |
CN109542623A (en) * | 2018-11-22 | 2019-03-29 | 中国人民解放军陆军工程大学 | The Optimization deployment method of virtual system in the server |
CN109542623B (en) * | 2018-11-22 | 2023-07-07 | 中国人民解放军陆军工程大学 | Optimized deployment method of virtual system in server |
CN109901932B (en) * | 2019-03-12 | 2023-04-07 | 东北大学 | Server integration method based on virtual machine |
CN109901932A (en) * | 2019-03-12 | 2019-06-18 | 东北大学 | A kind of Server Consolidation method based on virtual machine |
CN110321217A (en) * | 2019-03-25 | 2019-10-11 | 深圳大学 | A kind of cloud resource dispatching method, device, equipment and the storage medium of multiple target |
CN110008023B (en) * | 2019-03-26 | 2022-11-15 | 湖南农业大学 | Cloud computing system budget constraint random task scheduling method based on genetic algorithm |
CN110008023A (en) * | 2019-03-26 | 2019-07-12 | 湖南农业大学 | Cloud computing system budgetary restraints Random Task dispatching method based on genetic algorithm |
CN110363298A (en) * | 2019-07-19 | 2019-10-22 | 中建安装集团有限公司 | Intelligent adjustment of route alignment and gradient method based on evolution algorithm |
CN110363298B (en) * | 2019-07-19 | 2023-05-16 | 中建安装集团有限公司 | Intelligent line and slope adjusting method based on evolutionary algorithm |
CN110569589A (en) * | 2019-08-30 | 2019-12-13 | 西南交通大学 | Aisle arrangement method |
CN112578436A (en) * | 2019-09-27 | 2021-03-30 | 中国石油化工股份有限公司 | Wavelet extraction method and system |
CN112651538A (en) * | 2019-10-11 | 2021-04-13 | 阿里巴巴集团控股有限公司 | Method and device for configuring rack |
CN111651434A (en) * | 2019-10-22 | 2020-09-11 | 摩登汽车有限公司 | Millimeter wave radar data processing method, device and equipment based on genetic algorithm |
CN111651434B (en) * | 2019-10-22 | 2023-10-24 | 摩登汽车有限公司 | Millimeter wave radar data processing method, device and equipment based on genetic algorithm |
CN110851257A (en) * | 2019-11-07 | 2020-02-28 | 中国石油大学(华东) | Genetic and differential hybrid evolution cloud computing task scheduling algorithm based on early-stage catastrophe strategy |
CN112787833A (en) * | 2019-11-07 | 2021-05-11 | 中国电信股份有限公司 | Method and device for deploying CDN (content delivery network) server |
CN112787833B (en) * | 2019-11-07 | 2022-08-12 | 中国电信股份有限公司 | Method and device for deploying CDN (content delivery network) server |
CN110851277A (en) * | 2019-11-08 | 2020-02-28 | 中国石油大学(华东) | Task scheduling strategy based on edge cloud cooperation in augmented reality scene |
CN111079987A (en) * | 2019-11-28 | 2020-04-28 | 电子科技大学 | Semiconductor workshop production scheduling method based on genetic algorithm |
CN110942811B (en) * | 2019-12-14 | 2023-05-02 | 华东理工大学 | Method for designing catalyst in surface catalytic reaction based on genetic algorithm |
CN110942811A (en) * | 2019-12-14 | 2020-03-31 | 华东理工大学 | Design method of catalyst in surface catalytic reaction based on genetic algorithm |
CN111126707B (en) * | 2019-12-26 | 2023-10-27 | 华自科技股份有限公司 | Energy consumption equation construction and energy consumption prediction method and device |
CN111126707A (en) * | 2019-12-26 | 2020-05-08 | 华自科技股份有限公司 | Energy consumption equation construction and energy consumption prediction method and device |
CN113132445A (en) * | 2020-01-10 | 2021-07-16 | 阿里巴巴集团控股有限公司 | Resource scheduling method, device, network system and storage medium |
CN111291888A (en) * | 2020-01-21 | 2020-06-16 | 西安科技大学 | Scheduling optimization method for airport special vehicles |
CN111325648A (en) * | 2020-02-20 | 2020-06-23 | 杭州电子科技大学 | Aging service combination optimization method and system based on hesitation ambiguity |
CN111325481B (en) * | 2020-03-16 | 2023-02-24 | 郑州大学 | Intelligent automatic container terminal energy-saving comprehensive scheduling method |
CN111325481A (en) * | 2020-03-16 | 2020-06-23 | 郑州大学 | Intelligent automatic container terminal energy-saving comprehensive scheduling method |
CN111679907B (en) * | 2020-05-14 | 2023-04-18 | 西安交通大学 | Cloud computing task scheduling method |
CN111679907A (en) * | 2020-05-14 | 2020-09-18 | 西安交通大学 | Cloud computing task scheduling method |
CN111611748B (en) * | 2020-05-25 | 2023-05-26 | 上海大学 | Data-driven material reverse design method and system |
CN111611748A (en) * | 2020-05-25 | 2020-09-01 | 上海大学 | Data-driven material reverse design method and system |
CN111652373A (en) * | 2020-05-29 | 2020-09-11 | 济南浪潮高新科技投资发展有限公司 | Task scheduling method, device, equipment and storage medium of cloud robot |
CN111652373B (en) * | 2020-05-29 | 2023-10-31 | 山东浪潮科学研究院有限公司 | Task scheduling method, device and equipment of cloud robot and storage medium |
CN111724044A (en) * | 2020-05-29 | 2020-09-29 | 浙江汉脑数码科技有限公司 | Intelligent production scheduling system based on genetic algorithm |
CN111626516A (en) * | 2020-05-30 | 2020-09-04 | 湖南科技大学 | Double-deep-position four-way shuttle system order ordering optimization method considering goods reversing strategy |
CN111737120A (en) * | 2020-06-15 | 2020-10-02 | 河南工业大学 | Software defect repairing method and device |
CN111737120B (en) * | 2020-06-15 | 2022-09-23 | 河南工业大学 | Software defect repairing method and device |
CN111985647A (en) * | 2020-07-21 | 2020-11-24 | 西安理工大学 | Printing bookbinding job scheduling optimization method based on genetic algorithm |
CN112148443A (en) * | 2020-09-02 | 2020-12-29 | 安徽中科龙安科技股份有限公司 | Task optimization scheduling method and system of Loongson big data all-in-one machine |
CN112272102A (en) * | 2020-09-11 | 2021-01-26 | 北京工业大学 | Method and device for unloading and scheduling edge network service |
CN112272102B (en) * | 2020-09-11 | 2023-07-14 | 北京工业大学 | Method and device for unloading and scheduling edge network service |
CN112181655A (en) * | 2020-09-30 | 2021-01-05 | 杭州电子科技大学 | Hybrid genetic algorithm-based calculation unloading method in mobile edge calculation |
CN112214320A (en) * | 2020-10-10 | 2021-01-12 | 福州大学 | PSO-GA-based cloud service resource self-adaptive distribution method facing load time window |
CN112231117B (en) * | 2020-10-13 | 2023-05-09 | 山东大学 | Cloud robot service selection method and system based on dynamic vector hybrid genetic algorithm |
CN112231117A (en) * | 2020-10-13 | 2021-01-15 | 山东大学 | Cloud robot service selection method and system based on dynamic vector hybrid genetic algorithm |
CN112272381B (en) * | 2020-10-22 | 2022-08-30 | 大连大学 | Satellite network task deployment method and system |
CN112272381A (en) * | 2020-10-22 | 2021-01-26 | 大连大学 | Satellite network task deployment method and system |
CN112256402B (en) * | 2020-10-30 | 2024-03-29 | 深圳供电局有限公司 | Cloud platform data center resource prediction and scheduling method and system |
CN114443258B (en) * | 2020-10-30 | 2023-02-28 | 中国移动通信集团浙江有限公司 | Resource scheduling method, device, equipment and storage medium for virtual machine |
CN112256402A (en) * | 2020-10-30 | 2021-01-22 | 深圳供电局有限公司 | Cloud platform data center resource prediction and scheduling method and system |
CN114443258A (en) * | 2020-10-30 | 2022-05-06 | 中国移动通信集团浙江有限公司 | Resource scheduling method, device and equipment for virtual machine and storage medium |
CN112486651A (en) * | 2020-11-30 | 2021-03-12 | 中国电子科技集团公司第十五研究所 | Cloud test platform task scheduling method based on improved genetic algorithm |
CN112731378A (en) * | 2020-12-15 | 2021-04-30 | 郑州大学 | Method for acquiring dielectric constant by ant colony genetic combination, radar detection method and system |
CN112731378B (en) * | 2020-12-15 | 2024-03-26 | 郑州大学 | Ant colony genetic combination dielectric constant acquisition method, radar detection method and radar detection system |
CN112749776A (en) * | 2021-01-12 | 2021-05-04 | 南京信息工程大学 | Job shop scheduling method based on improved hybrid genetic algorithm |
CN112749776B (en) * | 2021-01-12 | 2023-08-15 | 南京信息工程大学 | Job shop scheduling method based on improved hybrid genetic algorithm |
CN112949195A (en) * | 2021-03-10 | 2021-06-11 | 国网吉林省电力有限公司信息通信公司 | Multi-target task scheduling method for power cloud data center |
CN112884368B (en) * | 2021-03-23 | 2022-11-01 | 合肥工业大学 | Multi-target scheduling method and system for minimizing delivery time and delay of high-end equipment |
CN112884368A (en) * | 2021-03-23 | 2021-06-01 | 合肥工业大学 | Multi-target scheduling method and system for minimizing delivery time and delay of high-end equipment |
CN113487142A (en) * | 2021-06-15 | 2021-10-08 | 昆山翦统智能科技有限公司 | Evolution optimization method and system for E-government performance assessment management |
CN113641471A (en) * | 2021-07-30 | 2021-11-12 | 平安科技(深圳)有限公司 | Soft load scheduling method, device, equipment and medium based on genetic algorithm model |
CN113641471B (en) * | 2021-07-30 | 2024-02-02 | 平安科技(深圳)有限公司 | Soft load scheduling method, device, equipment and medium based on genetic algorithm model |
CN114189914A (en) * | 2021-11-04 | 2022-03-15 | 国网江苏省电力有限公司南京供电分公司 | WiFi6 optimized switching decision method and system based on service priority |
CN114189914B (en) * | 2021-11-04 | 2023-12-19 | 国网江苏省电力有限公司南京供电分公司 | WiFi6 optimization switching decision method and system based on service priority |
CN114285880B (en) * | 2021-12-27 | 2023-06-23 | 电子科技大学长三角研究院(衢州) | Intelligent method for lightweight equipment of Internet of things based on enhanced twinning |
CN114285880A (en) * | 2021-12-27 | 2022-04-05 | 电子科技大学长三角研究院(衢州) | Internet of things lightweight equipment intelligentization method based on enhanced twinning |
CN114528088A (en) * | 2022-04-25 | 2022-05-24 | 北京航天驭星科技有限公司 | Satellite detection task processing method and device, electronic equipment and storage medium |
CN114840326A (en) * | 2022-06-01 | 2022-08-02 | 中国电信股份有限公司 | Resource scheduling method and device, electronic equipment and storage medium |
CN115185655A (en) * | 2022-06-23 | 2022-10-14 | 郑州轻工业大学 | Genetic task scheduling method based on gene frequency improvement |
CN115733839A (en) * | 2022-11-14 | 2023-03-03 | 中国电子科技集团公司第五十四研究所 | Bandwidth self-adaptive data chain intelligent distribution strategy optimization method |
CN115733839B (en) * | 2022-11-14 | 2023-09-12 | 中国电子科技集团公司第五十四研究所 | Bandwidth-adaptive data link intelligent distribution strategy optimization method |
CN116307296B (en) * | 2023-05-22 | 2023-09-29 | 南京航空航天大学 | Cloud resource optimization configuration method |
CN116307296A (en) * | 2023-05-22 | 2023-06-23 | 南京航空航天大学 | Cloud resource optimization configuration method |
CN116922402A (en) * | 2023-09-19 | 2023-10-24 | 杭州芯控智能科技有限公司 | Genetic algorithm-based multi-mechanical arm multi-task distribution method, system and medium |
CN116968037B (en) * | 2023-09-21 | 2024-01-23 | 杭州芯控智能科技有限公司 | Multi-mechanical-arm cooperative task scheduling method |
CN116968037A (en) * | 2023-09-21 | 2023-10-31 | 杭州芯控智能科技有限公司 | Multi-mechanical-arm cooperative task scheduling method |
CN117290077A (en) * | 2023-11-24 | 2023-12-26 | 深圳市飞梵实业有限公司 | Medical platform cloud computing task scheduling method and system based on genetic algorithm |
CN117290077B (en) * | 2023-11-24 | 2024-02-23 | 深圳市飞梵实业有限公司 | Medical platform cloud computing task scheduling method and system based on genetic algorithm |
Also Published As
Publication number | Publication date |
---|---|
CN104811491A (en) | 2015-07-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2016165392A1 (en) | Genetic algorithm-based cloud computing resource scheduling method | |
Pang et al. | An EDA-GA hybrid algorithm for multi-objective task scheduling in cloud computing | |
Liu et al. | Deadline‐constrained coevolutionary genetic algorithm for scientific workflow scheduling in cloud computing | |
Wang et al. | Load balancing task scheduling based on genetic algorithm in cloud computing | |
WO2020082973A1 (en) | Load prediction method and apparatus based on neural network | |
CN108469983B (en) | Particle swarm algorithm-based virtual machine deployment method in cloud environment | |
Zhang et al. | Network-aware virtual machine migration in an overcommitted cloud | |
US20110307899A1 (en) | Computing cluster performance simulation using a genetic algorithm solution | |
US11055139B2 (en) | Smart accelerator allocation and reclamation for deep learning jobs in a computing cluster | |
Adrian et al. | Analysis of K-means algorithm for VM allocation in cloud computing | |
CN111176784A (en) | Virtual machine integration method based on extreme learning machine and ant colony system | |
Huang et al. | Computation offloading for multimedia workflows with deadline constraints in cloudlet-based mobile cloud | |
Han et al. | An adaptive scheduling algorithm for heterogeneous Hadoop systems | |
Jiang et al. | Network-aware virtual machine migration based on gene aggregation genetic algorithm | |
Zhang et al. | Task scheduling of cloud computing based on Improved CHC algorithm | |
Hu et al. | Research of scheduling strategy on OpenStack | |
TW201734831A (en) | Method of assigning application to assigned service cluster and device | |
Chen et al. | A cloud computing resource scheduling scheme based on estimation of distribution algorithm | |
Chen et al. | A cost minimization data allocation algorithm for dynamic datacenter resizing | |
CN110908800B (en) | Scheduling optimization method based on scheduling entropy in edge cloud collaborative environment | |
Wang et al. | Spark load balancing strategy optimization based on internet of things | |
Liu | A Programming Model for the Cloud Platform | |
Ye | Optimization of resource scheduling based on genetic algorithm in cloud computing environment | |
Li et al. | Efficient fog node resource allocation algorithm based on taboo genetic algorithm | |
Yao et al. | An energy efficient virtual machine placement algorithm based on graph partitioning in cloud data center |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 15889088 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 05/02/2018) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 15889088 Country of ref document: EP Kind code of ref document: A1 |