NAS Experience with the Cray X1

A Cray X1 computer system was installed at the NASA Advanced Supercomputing (NAS) facility at NASA Ames Research Center in 2004. An evaluation study of this unique high performance computing (HPC) architecture, from the standpoints of processor and system performance, ease of use, and production computing readiness tailored to the needs of the NAS scientific community, was recently completed. The results of this study are described in this article. The evaluation included system performance and characterization using a subset of the HPC Challenge benchmarks and NAS Parallel Benchmarks, as well as detailed performance on scientific applications from the computational fluid dynamics (CFD) and Earth science disciplines that represent a major portion of the computational workload at NAS. Performance results are discussed and compared with the main HPC platform at NAS — the SGI Altix.

[1]  D. Johnson,et al.  Precipitation Processes During ARM ( 1997 ) , TOGA COARE ( 1992 ) , GATE ( 1974 ) , SCSMEX ( 1998 ) , and KWAJEX ( 1999 ) : Consistent 2 D and 3 D Cloud Resolving Model Simulations , 2003 .

[2]  R. Vanderwijngaart,et al.  NAS Parallel Benchmarks, Multi-Zone Versions , 2003 .

[3]  Rupak Biswas,et al.  Unstructured adaptive meshes: bad for your memory? , 2005 .

[4]  Michael Frumkin,et al.  Implementation of NAS Parallel Benchmarks in High Performance Fortran , 2000 .

[5]  Haoqiang Jin,et al.  - 1-Experience on the Parallelization of a Cloud Modeling Code Using Computer-Aided Tools , 2003 .

[6]  Man Mohan Rai,et al.  Multipassage three-dimensional Navier-Stokes simulation of turbine rotor-stator interaction , 1993 .

[7]  Man Mohan Rai,et al.  Three-dimensional Navier-Stokes simulations of turbine rotor-stator interaction , 1988 .

[8]  W-K. Tao.  Precipitation Processes developed during ARM (1997), TOGA COARE (1992), GATE (1974), SCSMEX (1998), and KWAJEX (1999), Consistent 2D, semi-3D and 3D Cloud Resolving Model Simulations , 2003 .

[9]  J. Taft.  PERFORMANCE OF THE OVERFLOW-MLP CFD CODE ON THE NASA AMES 512 CPU ORIGIN SYSTEM , 2000 .

[10]  Chung-Lin Shie,et al.  Implementation of a Message Passing Interface into a Cloud-Resolving Model for Massively Parallel Computing , 2004 .

[11]  David H. Bailey,et al.  The Nas Parallel Benchmarks , 1991, Int. J. High Perform. Comput. Appl..

[12]  William Chan,et al.  PARALLEL UNSTEADY TURBOPUMP SIMULATIONS FOR LIQUID ROCKET ENGINES , 2000, ACM/IEEE SC 2000 Conference (SC'00).

[13]  Stuart E. Rogers,et al.  Incompressible Navier-Stokes Solvers in Primative Variables and their Applications to Steady and Unsteady Flow Simulations , 2002 .