We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Reliable Task Scheduling Based on Resource Availability in Cloud Computing.
- Authors
S., Chitra; C. S. R., Prashanth
- Abstract
With ever increasing scale of cloud computing systems, to execute compute intensive parallel applications, in an environment where cloud resource availability is dynamic, effective task scheduling with reliability is an important factor to be considered. Virtual machines provided by current cloud infrastructures do not exhibit a stable performance due to network connectivity as well as computational nodes and may have a significant impact while scheduling workflows on clouds. Since the resources are not dedicated and can be used by other users simultaneously, there are load variations on resources, resulting in fluctuations in resource availability that affects the schedules. In this paper, a new reliable task scheduling algorithm based on resource availability is proposed, where task scheduling decision is based not only on task ready time but the processor available time slots that affect finish time of task. The main objective of the proposed technique is to reduce the overall execution time without increasing resource need. When algorithms that assume full resource availability are executed in that scenario, resource non-available slots contribute to delays thus increasing makespan. The proposed algorithm takes into consideration the resource availability factor during the prediction interval of application execution, while mapping tasks to resources during scheduling decisions, thus increasing reliability.
- Publication
International Journal of Advanced Research in Computer Science, 2017, Vol 8, Issue 5, p2747
- ISSN
0976-5697
- Publication type
Academic Journal