Do specialized distributed frameworks for bioinformatics applications obtain better performance over generic ones?

Kanak Mahadik, Wei Tang, Saurabh Bagchi, Folker Meyer

Abstract: The most popular approach to tackle the data deluge due to high throughput sequencing instruments is parallelizing applications and distributing the large datasets across cluster of computers to achieve scalability and performance. Hadoop is a generic and Shock-AWE is a customized platform for genomic data for development of such applications. In this work we compare and contrast performance of protein similarity search application based on these platforms.