A cost-effective approach to improving performance of big genomic data analyses in clouds

christopher smowton,andoena balla,demetris antoniades,crispin j miller,george pallis,marios d dikaiakos,wei xing

DOI: https://doi.org/10.1016/j.future.2015.11.011

IF: 7.307

2017-01-01

Future Generation Computer Systems

Abstract:With the rapidly growing demand for DNA analysis, the need for storing and processing large-scale genome data has presented significant challenges. This paper describes how the Genome Analysis Toolkit (GATK) can be deployed to an elastic cloud, and defines policy to drive elastic scaling of the application. We extensively analyse the GATK to expose opportunities for resource elasticity, demonstrate that it can be practically deployed at scale in a cloud environment, and demonstrate that applying elastic scaling improves the performance to cost tradeoff achieved in a simulated environment.