Army High Performance Computing Research Center

AHPCRC Technical Areas

SPSA iterations

TA4: HPC Enabling Technologies and Advanced Algorithms

Lead: Pat Hanrahan (Stanford)

As computing applications become more demanding, computers and the programs that run on them must evolve as well. This is especially true for high performance computing, where power consumption levels, computing resource usage, and program portability between platforms are essential to the effective use of these resources. Capabilities developed for AHPCRC point the way toward wider application as multicore processors and parallel programming become common in the commercial marketplace.

Graphic: Simultaneous Perturbation Stochastic Approximation, a method for global optimization in systems with local minima.

Projects and People
4–1: Stream Programming for High Performance Computing William Dally (Stanford) Alex Aiken (Stanford) Pat Hanrahan (Stanford)
4–2: Flexible Architecture Research Machine (FARM): Transactional Memory Kunle Olukotun (Stanford) Christos Kozyrakis (Stanford)
4–3: Simulation & Modeling to Enhance the Performance of Systems of Multicore Processors Pat Teller (UTEP) Jeanine Cook (NMSU)
4–4: High Performance Optimization Library Walter Murray (Stanford) Michael Saunders (Stanford)
4–5: PFMMPACK Parallel Fast Multipole Method Library Eric Darve (Stanford)
4–6: Hybrid Optimization Schemes for Parameter Estimation Problems Miguel Argáez (UTEP) Leticia Velázquez (UTEP)

Projects and People

4–1: Stream Programming for High Performance Computing