A Study on Dynamic Resource Allocation and Load Balancing using Gossip Protocol in Large Scale Cloud Environment

Abstract

The new emerging cloud computing technology is providing various services to users and
organizations. Now we are focusing on Infrastructure as a service which allows customers to use resource
multiplexing through virtualization technology. Cloud computing applications are developed using Map Reduce
programming. Cloud computing distributed file system nodes perform functionality like computing as well as
storage in this a file is partitioned into a number of chunks allocated in distinct nodes. based on that Map Reduce
tasks can be performed in parallel over the nodes but in a cloud computing environment, failure is the common
means nodes may be upgraded, replaced, and added in the system. Files can also be dynamically created, deleted,
and appended. This results in load imbalance in a distributed file system; that is, the file chunks are not
distributed as uniformly among the nodes. In this project introducing distributed load rebalancing algorithm
which eliminated load imbalance problem and proposing gossip protocol which maintains dynamic resource
management for large scale cloud. It dynamically maximizing the cloud utility under CPU and memory
constraints. In this paper, various resource allocation strategies and their challenges are discussed in detail. It is
believed that this paper would benefit both cloud users and researchers in overcoming the challenges faced.