Hybrid symbiotic organisms search optimization algorithm for scheduling of tasks on cloud computing environment
Cloud computing has attracted significant attention from research community because of rapid migration rate of Information Technology services to its domain. Advances in virtualization technology has made cloud computing very popular as a result of easier deployment of application services. Tasks ar...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Published: |
Public Library of Science
2016
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/id/eprint/72454/ https://www.scopus.com/inward/record.uri?eid=2-s2.0-84977137887&doi=10.1371%2fjournal.pone.0158229&partnerID=40&md5=3d5c5d58d5c0c2614b58d1c0b1ae9012 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Cloud computing has attracted significant attention from research community because of rapid migration rate of Information Technology services to its domain. Advances in virtualization technology has made cloud computing very popular as a result of easier deployment of application services. Tasks are submitted to cloud datacenters to be processed on pay as you go fashion. Task scheduling is one the significant research challenges in cloud computing environment. The current formulation of task scheduling problems has been shown to be NP-complete, hence finding the exact solution especially for large problem sizes is intractable. The heterogeneous and dynamic feature of cloud resources makes optimum task scheduling non-trivial. Therefore, efficient task scheduling algorithms are required for optimum resource utilization. Symbiotic Organisms Search (SOS) has been shown to perform competitively with Particle Swarm Optimization (PSO). The aim of this study is to optimize task scheduling in cloud computing environment based on a proposed Simulated Annealing (SA) based SOS (SASOS) in order to improve the convergence rate and quality of solution of SOS. The SOS algorithm has a strong global exploration capability and uses fewer parameters. The systematic reasoning ability of SA is employed to find better solutions on local solution regions, hence, adding exploration ability to SOS. Also, a fitness function is proposed which takes into account the utilization level of virtual machines (VMs) which reduced makespan and degree of imbalance among VMs. CloudSim toolkit was used to evaluate the efficiency of the proposed method using both synthetic and standard workload. Results of simulation showed that hybrid SOS performs better than SOS in terms of convergence speed, response time, degree of imbalance, and makespan. |
|---|