We propose to purchase and operate a high performance computing facility, the Orchestra Supercomputer, as a shared resource for 25 investigators (24 of whom is currently NIH funded) housed in several different institutions in Boston. This request is a response to the dramatic increase in need for high- performance computing in all areas of biomedical research, ranging from genomics to epidemiology, models of social behavior, computational cell biology and clinical, medical and translational computing. The challenges of operating a general-access high-performance computational resource are many, ranging from the technical (e.g. managing data growth) to the social (e.g. sharing resources in a fair and balanced way). We have addressed many of these challenges in a pilot compute cluster that has been operational for 4 years, funded by seed funds from Harvard Medical School. It is now clear that a larger shared resource is both necessary and possible. Based on the needs of the community, we propose to create a new shared supercomputer that will support over 3,000 simultaneous jobs and would provide 190TB of local disk space and 120TB of high performance network scratch space. The supercomputer will be built and administered by a team with proven operational experience in delivering genuinely shared computational services to the local research community. To support the operations the School has committed over half a million dollars a year of operating funds in the form of staff, facilities and maintenance costs. The supercomputer will support a wide range of platforms and algorithms, as required by the varied needs of the biomedical research community, and will use """"""""fair share"""""""" algorithms to balance the workload across available nodes and ensure that all researchers have appropriate access to the resource. Most importantly, this facility will allow participating researchers to accomplish the goals of their NIH- funded research more rapidly and with substantially more scope for experimentation and creativity. The components required will be sourced from several US manufacturers, especially IBM, supporting jobs in the important areas of computer hardware and software. In addition, it will enable a substantial step forward in biomedical computing in the New England area, creating the core of a larger facility that will eventually serve many researchers across the region.

Public Health Relevance

Computation and information technology has an ever increasing importance in the work of biomedical researchers. As computers processors become faster and memory and storage space increase, scientists are able to run billions of computations per processor second and analyze terabytes of data to make advances in many different fields, such as genomics and personalized medicine;the study of diseases like HIV/AIDS and cancers;the use of computers, computer imaging, and robotics in hospitals and at the patient's bedside;and understanding how cells communicate and how embryos develop. This application would enable such advances by providing Boston area doctors and scientists with a supercomputer, built with parts from American manufacturers.

Agency
National Institute of Health (NIH)
Institute
National Center for Research Resources (NCRR)
Type
Biomedical Research Support Shared Instrumentation Grants (S10)
Project #
1S10RR028832-01
Application #
7837270
Study Section
Special Emphasis Panel (ZRG1-BST-M (30))
Program Officer
Birken, Steven
Project Start
2010-05-06
Project End
2011-11-05
Budget Start
2010-05-06
Budget End
2011-11-05
Support Year
1
Fiscal Year
2010
Total Cost
$3,697,235
Indirect Cost
Name
Harvard University
Department
Type
Schools of Medicine
DUNS #
047006379
City
Boston
State
MA
Country
United States
Zip Code
02115
Reshef, Yakir A; Finucane, Hilary K; Kelley, David R et al. (2018) Detecting genome-wide directional effects of transcription factor binding on polygenic disease risk. Nat Genet 50:1483-1493
Shrestha, Brikha R; Chia, Chester; Wu, Lorna et al. (2018) Sensory Neuron Diversity in the Inner Ear Is Shaped by Activity. Cell 174:1229-1246.e17
Aganj, Iman (2018) Automatic Verification of the Gradient Table in Diffusion-Weighted MRI Based on Fiber Continuity. Sci Rep 8:16541
Natarajan, Niranjana; Abbas, Yamen; Bryant, Donald M et al. (2018) Complement Receptor C5aR1 Plays an Evolutionarily Conserved Role in Successful Cardiac Regeneration. Circulation 137:2152-2165
Lodato, Michael A; Rodin, Rachel E; Bohrson, Craig L et al. (2018) Aging and neurodegeneration are associated with increased mutations in single human neurons. Science 359:555-559
Aganj, Iman; Harisinghani, Mukesh G; Weissleder, Ralph et al. (2018) Unsupervised Medical Image Segmentation Based on the Local Center of Mass. Sci Rep 8:13012
György, Bence; Lööv, Camilla; Zaborowski, Miko?aj P et al. (2018) CRISPR/Cas9 Mediated Disruption of the Swedish APP Allele as a Therapeutic Approach for Early-Onset Alzheimer's Disease. Mol Ther Nucleic Acids 11:429-440
Loh, Po-Ru; Genovese, Giulio; Handsaker, Robert E et al. (2018) Insights into clonal haematopoiesis from 8,342 mosaic chromosomal alterations. Nature 559:350-355
Aganj, Iman; Fischl, Bruce (2017) Multimodal Image Registration through Simultaneous Segmentation. IEEE Signal Process Lett 24:1661-1665
Aganj, Iman; Iglesias, Juan Eugenio; Reuter, Martin et al. (2017) Mid-space-independent deformable image registration. Neuroimage 152:158-170

Showing the most recent 10 out of 35 publications