Workload Division by Replication for High Availability in Cloud Computing: A Review

Abstract

Abstract
This paper presents a comprehensive study about the problem of increase in the workload on the data nodes in case of node deletion or error occurrence in connection with server. In cloud computing fast and concurrent access to data is required which may increase the probability of overloading of available nodes due to failure of any node. It is considered as a serious problem when there is a continuous increase in the number of user requests for the type of application on cloud. Workload division is a mean for distributing workloads across all the nodes or resources available by strategically replicating the content of failed node to available nodes. Availability of data and its performance degrades as load increases. Load balancing is necessary for efficient operations in distributed environments for maximizing throughput, minimizing response time and avoiding unnecessary overload on any node or resource in the network. Replication can be used for ensuring availability in presence of purposeful server dawn times for load balancing.

Keywords: Cloud computing, distributed file systems, workload, load balancing, metrics, replication

Cite this Article
Manisha, Sona Malhotra. Workload Division by Replication for High Availability in Cloud Computing-A Review. Journal of Computer Technology Application. 2015; 6(2): 25–28p.