Katie Antypas User Services Group
	Office: National Energy Research Scientific Computing Center (NERSC) Lawrence Berkeley National Laboratory Oakland Scientific Facility Room 264 415 20th Street, Oakland, CA 510 486-5575 Email: kantypas at lbl dot gov	Mailing Address: Lawrence Berkeley Laboratory MS 943R0256 One Cyclotron Road Berkeley, CA 94720

I work in the User Services Group at NERSC as a High Performance Computing Consultant helping scientists more effectively utilize NERSC's computing resources. On a daily basis this means helping users debug, run and optimize codes, trouble shoot and track machine performance. I'm currently co-leading the NERSC-6 implementation team, a project to deploy NERSC's first peta-flop system. The system, called "Hopper", after computational scientist Grace Hopper is a Cray system and will be delivered in 2 phases. The first phase, an XT5 arrived in the fall of 2009 and has 664 compute nodes each containing two 2.4 GHz AMD Opteron quad-core processors (5312 total cores). The second phase, arriving in 2010, will be combined with an upgraded phase 1 to create a system with over 150,000 compute cores.

I graduated from Wellesley College in 1999 with a major in Physics and minor in Economics. For a few years after college I worked as a strategy consultant at CSMG, a telecom consulting firm in Boston before returning to science and enrolling in a Computer Science Masters program at the University of Chicago. After graduation I worked as a scientific programmer at the Flash Center at the U of C helping to support the FLASH code, a parallel adaptive mesh astrophysics code. In the fall of 2006, I returned to the Bay Area to work at NERSC.

Projects and Interests

• Utilization and efficiency of supercomputers
• Multicore programming models
• I/O benchmarking and performance; Parallel I/O strategies
• Parallel Debugging Tools, DDT and Totalview
• Testing and tracking machine performance with HPC benchmarks

Presentations and Tutorials

• Hopper: XT5 at NERSC, Joint XT5 Workshop with NERSC, OLCF and NICS, Berkeley, CA, Feb, 2010.
• NERSC: Delivering High-End Scientific Computing to the Nation's Research Community, Educause Conference, Denver CO, Nov 5, 2009.
• NERSC Overview, CSCADS Workshop on Petascale Applications and Performance Strategies, Snowbird, Utah, July 14, 2008.
• IO Monitoring on Franklin, Cray Technical Workshop, San Francisco, February 27, 2008.
• SVN Tutorial, NERSC Brown Bag Lunch Series (with Shreyas Cholia), October 10, 2007
• Franklin IO Best Practices, NERSC Users Group Meeting, Berkeley Lab, Oakland Scientific Facility, September 19, 2007.
• A New Parallel Debugger for NERSC: DDT, NERSC Users Group Meeting, Berkeley Lab, Oakland Scientific Facility, September 17, 2007.
• Parallel IO Benchmarking on GPFS, ScicomP 13, Garching by Munich, Germany, July 18, 2007

Reports

• DDT as a Parallel Debugging Alternative to Totalview , May 2007.

Published Papers and Conference Proceedings

• Extensible Component Based Architecture for FLASH: A Massively Parallel, Multiphysics Simulation Code. A. Dubey, K. Antypas, M.K. Ganapathy, L.B. Reid, K.M. Riley, D. Sheeler, A. Siegel, K. Weide., Parallel Computing (in press).
• MPI-I/O on Franklin XT4 System at NERSC. K. Antypas and A. Uselton., CUG 2009 Proceedings.
• Characterizing and Predicting the I/O Performance of HPC Applications Useing a Parameterized Synthetic Benchmark. H. Shan, K. Antypas, J.Shalf. IBM Journal of Research and Development, Supercomputing 2008 (SC08 Conference).
• Tera-scale Turbulence Computation on BG/L Using the FLASH3 Code. R. Fisher, S. Abarzhi, K. Antypas, S. M. Asida, A. C. Calder, F. Cattaneo, P. Constantin, A. Dubey, I. Foster, J. B. Gallagher, M. K. Ganapathy, C.C. Glendenin, L. Kadano, D.Q. Lamb, S. Needham, M. Papka, T. Plewa, L.B. Reid, P. Rich, K. Riley, and D. Sheeler. IBM Journal of Research and Development. Vol 52 March 2008.
• Scientic Applications on the Massively Parallel BG/L Machine. K. Antypas, A.C. Calder, A. Dubey, R. Fisher, M.K. Ganapathy, J.B. Gallagher, L.B. Reid, K. Riley, D. Sheeler, and N. Taylor, proc. Parallel and Distributed Processing Techniques and Applications (PDPTA '06)
• A Case Study of Verifying and Validating an Astrophysical Simulation Code. A. C. Calder, N. T. Taylor, K. Antypas, and D. Sheeler, in Numerical Modeling of Space Plasma Flows: Astronum-2006 ASP Conference Series, Volume 359, Proceedings of the Conference Held 26-30 March, 2006, in Palm Springs, California, USA. G.P. Zank and N.V. Pogorelov, eds. San Francisco: Astronomical Society of the Pacific., p. 119
• FLASH: Applications and Future. K.B. Antypas, A. C. Calder, A. Dubey, J. B. Gallagher, J. Joshi, D. Q. Lamb, T. Linde, E. Lusk, O. E. B. Messer, A. Mignone, H. Pan, M. Papka, F. Peng, T. Plewa, P. M. Ricker, K. Riley, D. Sheeler, A. Siegel, N. Taylor, J. W. Truran, N. Vladimirova, G. Weirs, D. Yu, Z. Zhang. in Parallel Computational Fluid Dynamics 2005: Theory and Applications, eds. A. Deane, G. Brenner, A. Ecer, D. R. Emerson, j. McDonough, J. Periaux, N. Satofuka, D. Tromeur-Dervout. Elsevier, Amsterdam, 2006, p. 325

---

Computing Sciences | Computational Research Division | ESnet

Page last modified: Mon, 12 Apr 2010 21:58:45 GMT Page URL: http://www.nersc.gov/homes/kantypas/ Web contact: webmaster@nersc.gov Computing questions: consult@nersc.gov Privacy and Security Notice