Security-aware And Fault-tolerant Workflow Scheduling In Mobile Edge Computing

With the development of mobile internet,mobile edge computing(MEC)has become a new mobile computing model.In this mode,computing resources,network control,and storage are deployed to the edge of the network.Mobile users can reduce latency and reduce energy consumption by migrating application tasks to the MEC.However,users who offload applications to the MEC are experiencing data security issues because any potential malicious attack can result in the loss or disclosure of private data.Security services are often used to secure data,but different security services create different security overheads.Therefore,how to use reasonable scheduling to minimize energy consumption and effectively deploy security services to meet the security requirements of applications has become an important part of workflow scheduling under MEC.Reliability requirements are one of the most important quality of service evaluation standard.Workflow scheduling should meet reliability requirements under MEC.However,in the MEC,VMs inevitably have software failures,which causes task execution failures to delay task completion time and affect service quality.To avoid a software failure that causes task execution to fail,we need to deploy a fault-tolerant policy to the system.However,an unreasonable fault-tolerant strategy will use more resources,which not only affects the quality of service but also increases the load on the MEC server.Therefore,how to choose the appropriate fault tolerance mechanism to maximize resource utilization becomes an important issue in workflow scheduling under MEC.Based on the above understanding,this paper studies the service workflow scheduling technology for security and energy consumption under the mobile edge computing environment.In addition,this paper studies the dynamic fault-tolerant scheduling algorithm of workflow in mobile edge computing environment for VM faults.The main contributions of this paper are as follows:(1)For the data security problem in MEC environment,this paper first constructs a new security cost model,including the relationship between data volume and security overhead,the relationship between multi-core CPU and security overhead,and the relationship between computation frequency and security overhead.The security overhead of the task can be quantitatively measured according to the parameters of the MEC.Then,a service workflow scheduling method(SEA)for security and energyawareness is proposed,which can minimize the energy consumption of mobile devices under the condition of meeting the risk rate and deadline of mobile applications.SEA is based on particle swarm optimization algorithm,which considers the scheduling position,confidentiality service and integrity service of the task.Finally,the simulation experiment proves the feasibility and effectiveness of the proposed algorithm.(2)For the VM fault problem in MEC environment,this paper studies the workflow fault tolerance problem in MEC environment,and proposes a dynamic faulttolerant scheduling(FATOS)algorithm,which can guarantee the successful execution of the task and maximize the MEC system in the event of VM failure.Resource utilization.This article first introduces two fault tolerance mechanisms(Delayed Execution Replication(R-DE)and Delayed Execution Checkpoint(C-DE))for handling VM failures.Then,the resource adjustment strategy is proposed to dynamically adjust the computing resource demand through the expansion and reduction of resources.Finally,a FATOS algorithm consisting of fault tolerance mechanism and resource adjustment strategy is proposed to schedule the workflow.Simulation experiments show that the FATOS algorithm outperforms other comparison algorithms in workflow rejection rate and resource utilization.