Speaker: Dr. Zhenman Fang, Ph.D.,
Computer Science Department,
University of California, Los Angeles
Time: 11:45 A.M. – 1 P.M. (Lunch will be serverd*)
Date: Apr 12, 2017
Location: 5628, Math Sciences Building
Abstract: Modern medical research is being transformed with the advance of next-generation sequencing technologies that dramatically reduce the cost of sequencing an individual human genome from $10M to $1,000 in the past decade. This sequencing technology is widely utilized in research and is transitioning into the clinic for applications such as sequencing cancer tumors. To be more cost-effective, genome sequencers obtain the sequence of billions of small fragments in the range of a few hundred nucleotides (called short reads) of a whole genome. Therefore, we have to reconstruct all these short reads back into a whole human genome for genome variant calling and cancer discovery. This poses great computational challenges: state-of-the-art DNA sequencing pipeline takes around a week on a 12-core CPU server.
In this talk, I will present various ways to accelerate the next-generation DNA sequencing pipeline done in our Center for Domain-Specific Computing (CDSC), with the final goal to reduce the computation time to a few hours. First, we scale out the computation by leveraging the power of a CPU cluster/datacenter. Second, we scale up the computation by developing hardware accelerators on commodity energy-efficient FPGAs that can be plugged into the CPU server. Moreover, we combine these two techniques and address the new challenges in FPGA-enabled datacenters. I will talk about our current status in accelerating the DNA sequencing pipeline and show some promising results we have achieved using the aforementioned techniques [FCCM 2015, HotCloud 2016, ACM SOCC 2016].
About Speaker: Dr. Zhenman Fang is an IDRE scholar and a postdoc in the Computer Science Department, UCLA, working with Prof. Jason Cong and Prof. Glenn Reinman. He is a member of the NSF/Intel funded multi-university Center for Domain-Specific Computing (CDSC) and the SRC/DARPA funded multi-university Center for Future Architectures Research (C-FAR). Zhenman received his PhD in June 2014 from Fudan University, China and spent the last 15 months of his PhD program visiting University of Minnesota at Twin Cities. Zhenman’s research lies at the boundary of big data workloads and systems, heterogeneous and energy-efficient accelerator-rich architectures and systems, and system-level design automation. He has published 10+ papers in top venues that span across computer architecture (HPCA, TACO, ICS), design automation (DAC, ICCAD, FCCM), and cloud computing (ACM SOCC). He received several awards, including a best paper nominee of HPCA 2017, a best demo award (3rd place) at the C-FAR center annual review. More details can be found in his personal website: https://sites.google.com/site/fangzhenman/.