Scheduling in CloudSim of Interdependent Tasks for SLA Design

Abstract

One of the most recent concepts in the framework of today's distributed systems is Cloud Computing. A very difficult problem that needs to be addressed is the management of the Cloud. When designing a Cloud scheduling strategy, the design trade-offs of the Cloud architecture should be evaluated. The easiest way to evaluate this infrastructure is to use a simulation tool (in this case CloudSim simulation toolkit). This article examines three different algorithms that consider the scheduling of tasks in Cloud, which are described as Directed Acyclic Graphs (DAGs) (interdependent tasks). The results of the scheduling algorithms are valuable because they are helpful for the design of the Cloud Computing scheduling algorithms and for the design of a Service Level Agreement (SLA) contract whose members are the Cloud Service Provider (CSP) and the Cloud Service Customers (CSCs). This paper shows that the selection of the scheduling algorithms is useful in the design of the SLA contract.

Bi-objective scheduling algorithm for scientific workflows on cloud computing platform with makespan and monetary cost minimization approach

“…Highest Level First with Estimated Times(HLFET) [15] Modified Critical Path(MCP) [16] Dynamic Critical Path(DCP) [17] Dynamic Level Scheduling(DLS) [18] Heterogeneous Earliest Finish Time(HEFT) [19] Critical Path On a Processor(CPOP) [19] Longest Dynamic Critical Path(LDCP) [20] Cost-effective fault Tolerance (CEFT) [21] QL-HEFT [22] Clustering and Scheduling System II (CASS II) [23] Dominant Sequence Clustering(DSC) [24] Clustering heuristic [21], [22], [25] Critical Path Fast Duplication(CPFD) [26] Duplication Scheduling Heuristic(DSH) [12], [13], [27], [28] Task Duplication-based Scheduling(TDS) [29] HEFT Task Duplication(HEFT-TD) [30] Lookahead HEFT-TD [30] Genetic Algorithm (GA) [31]- [34] Simulated Annealing(SA) [35]- [38] Particle Swarm Optimization(PSO) [39]- [44] Ant Colony Optimization(ACO) [45]- [51] Artificial Bee Colony (ABC) [52] Cuckoo Search algorithm(CS) [53] Task-Scheduling Algorithms…”

“…It preserves time complexity against other existing approaches, but it lacks to consider rented VMs' cost. Some other famous list schedulers are: Highest Level First with Estimated Times (HLFET) [15], Modified Critical Path (MCP) [16], Dynamic Critical Path (DCP) [17], Dynamic Level Scheduling (DLS) [18], and Longest Dynamic Critical Path (LDCP) [20] in which their concentration are mostly on critical path management of given DAGs.…”

“…After adding all tasks of set T to the list < T ask List > separately, it is turned to the second step which is placed in lines (15)(16)(17)(18)(19). In this section, there should be a random number assigned to the task that represents the number of duplicates for each task.…”

QoS-Aware Cost Minimization Strategy for AMI Applications in Smart Grid Using Cloud Computing

“…Although, in the cloud environment, horizontal (e.g., adding more virtual machines) and vertical (increase size of the CPU, storage, RAM, and bandwidth) scaling are employed by organizations to handle the growing demands, ensure uptime, and optimize the network performance. For simplicity, here, the cost of cloud resource [9,10] means the leasing of cloud resources or paying for the use of cloud services, e.g., database applications, which consumers pays to the cloud provider. In Smart Grid, QoS provision with agreed terms and conditions, i.e., service level agreements (SLAs), are set at the time of the smart meter connection and installation.…”

“…These traffic characteristics are listed in Table 4, and usually P L values are set by the network operator of the Utility provider. For clarity, P L assists in the creation of priority queues (qth) in order to avoid conflict of the queue allocation; that is, the queue contention of AMI application traffic requires priority-based transmission without delay and packet loss, whereas the priority metric (P k i, q ) computed in Equation (10) above assists in scheduling the AMI traffic in priority queues. In the proposed optimization model, four queues (Q = 1, .…”