Maurice Yarrow

Parallel and Distributed Computational Fluid Dynamics: Experimental Results and Challenges

This paper describes several results of parallel and distributed computing using a production flow solver program. A coarse grained parallelization based on clustering of discretization grids, combined with partitioning of large grids, for load balancing is presented. An assessment is given of its performance on tightly-coupled distributed and distributed-shared memory platforms using large-scale scientific problems. An experiment with this solver, adapted to a Wide Area Network environment, is also presented.

Finite-surface method for the Maxwell equations in generalized coordinates

A finite-surface technique for the Maxwell equations in generalized nonorthogonal coordinates is developed. It directly applies the integral Faradays and Amperes laws to faces of primary and secondary grid cells, respectively. The technique features an accurate treatment of matching conditions at a material interface, grid singularities, and radiation conditions at outer boundaries. 12 refs.

Multiprocessing on supercomputers for computational aerodynamics

Little use is made of multiple processors available on current supercomputers (computers with a theoretical peak performance capability equal to 100 MFLOPS or more) to improve turnaround time in computational aerodynamics. The productivity of a computer user is directly related to this turnaround time. In a time-sharing environment, such improvement in this speed is achieved when multiple processors are used efficiently to execute an algorithm. We apply the concept of mul tiple instructions and multiple data (MIMD) through multitasking via a strategy that requires relatively minor modifications to an existing code for a single processor. This approach maps the available memory to multiple processors, exploiting the C-Fortran-Unix interface. The existing code is mapped without the need for devel oping a new algorithm. The procedure for building a code utilizing this approach is automated with the Unix stream editor.

Finite-surface method for the Maxwell equations with corner singularities

The finite-surface method for the two-dimensional Maxwell equations in generalized coordinates is extended to treat perfect conductor boundaries with sharp corners. Known singular forms of the grid and the electromagnetic fields in the neighborhood of each corner are used to obtain accurate approximations to the surface and line integrals appearing in the method. Numerical results are presented for a harmonic plane wave incident on a finite flat plate. Comparisons with exact solutions show good agreement.

Predicting Price/Performance Trade-offs for Whitney: A Commodity Computing Cluster

We couple simple performance models with pricing to optimize the design of clusters built from commodity components for scientific computing. We apply this technique using the NAS Parallel Benchmarks as a representative workload. We develop models of the BT, LU, and SP benchmarks. The models consist of closed form expressions based on problem size, number of processors, and three measured quantities (single processor performance, network latency, and network bandwidth). These models predict benchmark performance to within 30%. This technique was used in the design of Whitney, a commodity computing cluster at NASA Ames Research Center. In particular, for systems costing less than

The NAS Parallel Benchmarks 2.0

1,000,000, the performance characteristics of Intel Pentium processors are better matched to the slower (and less expensive) Fast Ethernet, than to the faster (and more expensive) Myricom Myrinet.