Workload Balancing in a Multi-Cloud Environment: Challenges and Research Directions

Abstract

Single cloud provider can manage their cloud on their own individually to deliver services to the users based on the need and demand of the user anywhere, anytime. Multi-cloud technology has been proposed in recent years as a revolutionary solution where a user uses more than one cloud platform and each one of them delivers a specific application or service. Thus, the total workload can be distributed across multiple clouds. The collaboration of the cloud service providers in the heterogeneous multi-cloud environment and workload balancing among multiple clouds in a multi-cloud environment is very challenging. By balancing the load among the multiple clouds, multi-cloud technology avoids overloading at any cloud, reduces waiting time, improves response time of the requests, avoids service level agreement violations, results in effective utilization of resources, and optimizes power usage at each cloud. In the first part of this book chapter, multi-cloud architecture, working functionality of multi-cloud, managing multi-cloud environment such as balanced distribution of load among multiple clouds, power management in an optimized way at each cloud, monitoring of service level agreement, and satisfying service level agreement are presented. These parameters are crucial in the operation and effectiveness in the multi-cloud environment, and the influence of these parameters on the performance of the multi-cloud is also discussed. In the second part of the book chapter, current challenges in multi-cloud like balanced distribution of workload among multiple cloud, application and service deployment in a heterogeneous multi-cloud environment, use of resources and services from different cloud service providers, security in data storage, large scale data processing, and research directions toward the multi-cloud are presented. Finally, the simulators available for multi-cloud, applicability of simulators, and best simulators to solve research challenges in multi-cloud management are elaborated. This gives an insight in choosing the best and suitable simulator for solving different research issues.