Enabling Bioinformatics Discovery with High Performance Computing Solutions
Storing data is easy — the challenge is doing anything else

 

On Thursday, May 20, 2010, a group of approximately 50 life science researchers and IT professionals participated in a one-day workshop at the Broad Institute of the Massachusetts Institute of Technology (MIT) and Harvard University in Cambridge, MA. Tim Carroll, with Dell’s High Performance Computing Servers, Storage & Services Advanced Solutions Group, kicked off the event with a warm welcome for all participants and an overview of the day’s agenda, designed to examine the impact of various technologies and solutions upon the discovery process, as well as take a look at several exciting new developments in research computing.

Dell sponsored the workshop along with partners Intel, Isilon Systems, Qlogic and Scientific Computing, who also provided several opportunities throughout the day’s activities where attendees could be eligible to win prize drawings and giveaways.

The day’s first speaker was Glen Otero, Ph.D., an HPC product technologist for HPC life sciences at Dell. Otero spoke on new directions in bioinformatics, including new products, services and partnerships Dell is developing to simplify research computing for life scientists. His talk included a look at the need for robust storage in HPC environments; hardware, such as the PowerEdge 6100; cloud computing; and research “community” initiatives.

Matthew Trunnell, Acting Director for Advanced IT at the Broad Institute, then shared his perspective on research computing, as well as ways in which the Broad Institute is tackling today’s storage challenges in order to deliver on its mission to discover, develop and optimize the critical technologies needed to obtain and analyze the massive amounts of genomic data being generated by scientists at the institute. The Broad’s adoption of second-generation sequencing technologies over the past three years has driven a 30-fold growth in the size of its data repositories, placing new pressures on IT infrastructure, motivating different approaches to data analysis, and demanding a new level of attention to issues of data management. Trunnell described the recent evolution of the research computing infrastructure at the Broad and examined the various impacts of this new scale of data, including setting storage purchase priorities, data protection and key data management issues.

Goran Ceric, the manager of scientific computing systems for the Janelia Farm Research Campus of the Howard Hughes Medical Institute (HHMI), spoke on “The Science of IT” at Janelia Farm, a biomedical research center where scientists from diverse disciplines use emerging and innovative technologies to pursue biology’s most challenging problems. The center’s research goals include identification of general principles guiding how information is processed by neuronal circuits, in addition to the development of new imaging technologies and computational methods for image data analysis. Ceric covered how HHMI’s IT resources are organized and looked at the institute’s infrastructure. He discussed examples of several scientific projects that heavily utilize the infrastructure and addressed challenges, such as archiving and digital preservation, digital asset management and how to drive cultural change.

Clem T. Cole, an HPC architect with Intel Cluster Ready, spoke on “The Pursuit of EveryScale Computing.” Cole examined current trends in the HPC market and looked in detail at what the U.S. Council on Competitiveness – a non-profit organization whose goal is to increase U.S. economic competitiveness in the global marketplace by bringing together business, labor, academic and government leaders to evaluate economic challenges – has called the “missing middle.” He explained why this “missing middle” is the key area of HPC opportunity that has not yet been realized.

The day wrapped up with an animated question and answer session that included Otero, Cole, Trunnel and Ceric, as well as Tom Manganaro of Isilon Systems and Lloyd Dickman, Qlogic’s CTO for Infiniband products in their Network Systems Group.

• For further information: www.DellHPCSolutions.com, www.ScientificComputing.com