Font Size: a A A

Research On User-Oriented Multi-Objective Cloud Workflow Scheduling

Posted on:2018-01-07Degree:MasterType:Thesis
Country:ChinaCandidate:S H LiuFull Text:PDF
GTID:2348330536962012Subject:Information management and e-government
Abstract/Summary:PDF Full Text Request
Cloud computing is able to deliver large amount of computing resources on demand,and it has become one of the most effective ways to implement large-scale computationally intensive applications.In a cloud computing environment,applications typically involve workflows.Therefore,optimized workflow scheduling can greatly improve the overall performance of cloud computing.However,existing studies on cloud workflow scheduling usually consider one single objective only and effective methods to solve scheduling problems with four or more objectives still lack.And most of studies consider fault-tolerance by taking reliability as an objective to obtain a highly reliable schedule while the fault-tolerance mechanism for task execution failure are seldom researched.Since it is common for task failure during execution caused by hardware and software fault,the effective strategy for rescheduling failed tasks are necessary.To address the above issue,this work mainly includes the following two aspects:(1)A new cloud workflow scheduling model is formulated.to solve this model,an improved KnEA is proposed.the proposed model considers four objectives,namely,minimization of makespan,minimization of the average execution time of all workflow instances,maximization of reliability,and minimization of the cost of workflow execution simultaneously.To solve this four-objective scheduling problem,an improved knee point driven evolutionary algorithm is proposed.Extensive experimental results demonstrate that the improved algorithm outperforms existing popular many-objective evolutionary algorithms in most experimental scenarios studied in this work,in particular when there is sufficiently large amount of computing resource supply and the time for scheduling is limited.(2)A rescheduling method is proposed aiming at task failure caused by task exception or software fault.Firstly,based on the computation resource(virtual machine)idle time interval,a cloud workflow rescheduling model is proposed aiming at task failure.This model considers three objectives including the completion time and cost of rescheduled tasks,the reliability of reschedule workflow instance.Then,using multi-criteria decision making and method Combining Multi-objective Optimization with Multi-criteria Decision to deal with scenarios with different number of reschedule plans,which results from varying number of rescheduled tasks and available virtual machines.Finally,an example is given to illustrate the rescheduling mechanism.This paper presents a user-oriented fault-tolerant cloud workflow scheduling method,which can further meet the diversity of cloud workflow scheduling requirements and improve the quality of cloud computing services.Thus,it can enrich the theoretical method system of cloud workflow scheduling,and has good practical application prospect.
Keywords/Search Tags:Cloud Computing, Workflow Scheduling, Many-objective optimization problems, Rescheduling
PDF Full Text Request
Related items