Resource reliability using fault tolerance in cloud computing
Authors:
- 16 March 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 65-71
Abstract
Cloud computing offers variety of services from software instance to resource provisioning. As the user demands increase, there is a necessity to enhance cloud offerings. But still in some cases fault-tolerance is the major challenge for cloud environment. Multiple request to access the same server sometime leads to server over loaded and may increase faults and cause unreliability for the server. Few fault-tolerance techniques like self-healing, job migration, static load balancing and replication are existed but they are not fully reliable and effective for cloud environment. In this paper, we propose pro-active approach for fault-tolerance based on Processing power, Memory and Network parameters to increase resource reliability. Through this approach, we first calculate the reliability of each Virtual Machine (VM) based on success rate of task execution and then schedule the task on highly reliable VM. This approach provides comparatively good results for the VM reliability and system stability.Keywords
This publication has 13 references indexed in Scilit:
- Characterizing and profiling scientific workflowsFuture Generation Computer Systems, 2013
- Deadline-constrained workflow scheduling algorithms for Infrastructure as a Service CloudsFuture Generation Computer Systems, 2013
- Cost-Driven Scheduling of Grid Workflows Using Partial Critical PathsIEEE Transactions on Parallel and Distributed Systems, 2011
- The Characteristics of Cloud Computing2010
- CloudSim: a toolkit for modeling and simulation of cloud computing environments and evaluation of resource provisioning algorithmsSoftware: Practice and Experience, 2010
- The Failure Trace Archive: Enabling Comparative Analysis of Failures in Diverse Distributed Systems2010
- Evaluation and Optimization of the Robustness of DAG Schedules in Heterogeneous EnvironmentsIEEE Transactions on Parallel and Distributed Systems, 2009
- Cloud Computing and Grid Computing 360-Degree Compared2008
- Enriching network security analysis with time travel2008
- Robust task scheduling in non-deterministic heterogeneous computing systems2006