Energy Efficiency Problems In IaaS Level Of Cloud

In recent years,Cloud has become the dominant platform for scholars and service providers to deploy their computing applications.Benefiting from the advantages of cloud platforms,such as dynamic scalability,on-demand deployment,high reliability,and cost-effective,they have attracted many users,scholars deploy large-scale scientific computing programs on the cloud,and service providers deploy a variety of services,such as Web services and Mail services,on the cloud.With the development and innovation of technology,such as the improvement of host resource balancing and cyclical load in services,data centers have also brought up some important issues that need to be resolved in new scenarios,such as cost issues,fault tolerance issues,power consumption issues and stability issues,which profoundly affect the interests of the users and the cloud providers.Since IaaS is the main model for service delivery on cloud platforms,this paper explores these issues in depth based on the IaaS layer.In this paper,the research is carried out from two standpoints.On the one hand,the au-thors stand from the perspective of users who need to execute large-scale scientific program,and propose some algorithms and mechanisms to help them reduce the cost of using cloud platform and improve the reliability of large-scale program.On the other hand,the authors stand on the perspective of cloud providers to help them reduce the power consumption and improve the sta-bility of data centers.In the aforementioned research,this paper mainly achieves the following achievements:1.We address the problem of scheduling a scientific application on cloud platform from the perspective of users.First,we propose a Satisfiability Modulo Theories(SMT)based al-gorithm to schedule a scientific application on cloud platform,the SMT algorithm constructs the scheduling problem to first-order logic expressions and checks the expressions by solvers,which minimizes the number of Virtual Machine instances(VMs)allocated to the application.Furthermore,due to the hourly payment of cloud,we develop a heuristic algorithm called Mul-tiple Strategies Algorithm(MSA)which determines the minimum instance hours of a scientific application deployed on VMs.At last,we combine the proposed SMT based algorithm and the MSA to a framework named SMT-MSA,and compare it with other outstanding algorithms in experiments,the results show that,in most of cases,our algorithms reduce more cost than the other three methods which are HEFT,MSMD and IC-PCPD2.2.We present a novel task scheduling framework named cost minimization approach with DAG splitting method(COMSE)for minimizing the cost of running a deadline constrained large-scale scientific workflow.First,we provide comprehensive theoretical analyses on how to im-prove the utilization of a resource balanced multi-vCPU VM for running multiple tasks simulta-neously.Second,considering the balance between the parallelism and the topology of a work-flow,we simplify the DAG-based workflow,and based on the simplified DAG,a DAG splitting method is devised to preprocess the workflow.Third,since the cloud is charged by hours,we also design an exact algorithm to find the optimal operation pattern for a given schedule to make the consumed instance hours minimum,this algorithm is named as instance hours minimization by Dijkstra(TOID).Finally,by employing the DAG splitting method and the TOID,the COMSE schedules a deadline constrained large-scale scientific workflow on the multi-vCPU VMs,and incorporates two important objects:minimizing the computation cost and the communication cost.Our solution approach is evaluated through rigorous performance evaluation study using real-word workflows,and the results show that the proposed COMSE approach outperforms existing algorithms in terms of computation cost and communication cost.3.We present a novel fault tolerant framework named Fault Tolerance Algorithm using Selective Mirrored Tasks Method(FAUSIT)for the fault tolerant of running a big data appli-cation on cloud.First,we provide comprehensive theoretical analyses on how to improve the performance of fault tolerant for running a single task on a processor.Second,considering the balance between the parallelism and the topology of an application,we present a selective mir-rored task method.Finally,by employing the selective mirrored task method,the FAUSIT is desigened to improve the fault tolerant for DAG based application and incorporates two im-portant objects:minimizing the makaspan and the computation cost.Our solution approach is evaluated through rigorous performance evaluation study using real-word workflows,and the results show that the proposed FAUSIT approach outperforms existing algorithms in terms of makaspan and computation cost.4.We present a VM consolidation algorithm for predictable loads of VMs(VCPL)to reduce the live migration operations.First,we present a cyclic usage prediction(CUP)method to predict the load in a whole cycle of a VM.Then,we separate the VMs with cyclic load out from others and consolidate them to PMs by using VCPL to make sure each PM has a stable load.Thus,energy can be reduced by avoiding most of live migration operations.We evaluate our solution through simulations on real-world workloads,the results show that,66%of long-term VMs have cyclic loads and are predictable,by using VCPL,the live migration operations occurring on the PMs which accommodate those VMs can be reduced significantly than other solutions.