An Optimized Resource Allocation Approach for Data-Intensive Workloads Using Topology-Aware Resource Allocation

Abstract

This paper proposes an optimized resource allocation mechanism in Infrastructure-as-a-Service (IaaS)- based cloud systems. Performance of distributed data-intensive applications are impacted significantly as current IaaS systems are usually unaware of the hosted application's requirements and hence allocating resources independent of its needs. To address this resource allocation problem and to optimise the allocation, we enhance an architecture that adopts a "what if" methodology to guide allocation decisions taken by the IaaS. The architecture uses a prediction engine with a lightweight simulator to estimate the performance of a given resource allocation and an evolutionary algorithm that includes an evolution strategies algorithm and a genetic algorithm, to find an optimized solution in the large search space.