Abstract
With the increasing adoption of next generation sequencing technology in the medical practice, there is an increasing demand for faster data processing to gain immediate insights from the patient's genome. Due to the extensive amount of genomic information and its big data nature, data processing takes long time and delays are often experienced. In this paper, we show how to exploit in-memory platforms for big genomic data analysis, with focus on the variant analysis workflow. We will determine where different in-memory techniques are used in the workflow and explore different memory-based strategies to speed up the analysis. Our experiments show promising results and encourage further research in this area, especially with the rapid advancement in memory and SSD technologies.