Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

10

HPDC
1996
IEEE

favoriteEmaildiscussreport

121views Distributed And Parallel Com...» more HPDC 1996»

Shared Memory NUMA Programming on I-WAY

13 years 8 months ago

Shared Memory NUMA Programming on I-WAY

Download www.emsl.pnl.gov

The performance of the Global Array shared-memory nonuniform memory-access programming model is explored on the I-WAY, wide-area-network distributed supercomputer environment. The Global Array model is extended by introducing a concept of mirrored arrays. Latencies and bandwidths for remote memory access are studied, and the performance of a large application from computational chemistry is evaluated using both fully distributed and also mirrored arrays. Excellent performance can be obtained with mirroring if even modest (0.5 MB/s) network bandwidth is available.

Jarek Nieplocha, Robert J. Harrison

Real-time Traffic

Distributed And Parallel Computing | Global Array | Global Array Shared-memory | HPDC 1996 | Mirrored Arrays |

claim paper

Related Content

» A Programming Interface for NUMA SharedMemory Clusters

» Comparison of MPI benchmark programs on an SGI Altix ccNUMA shared memory machine

» Compiler Support for Array Distribution on NUMA Shared Memory Multiprocessors

» Garbage Collection for Multicore NUMA Machines

» MPC A Unified Parallel Runtime for Clusters of NUMA Machines

» Shared Memory Parallelization of the GROMOS96 Molecular Dynamics Code

» Algorithmic optimizations of a conjugate gradient solver on shared memory architectures

» A PRAMNUMA model of computation for addressing lowTLP workloads

» Reducing Ownership Overhead for LoadStore Sequences in CacheCoherent Multiprocessors

Post Info
More Details (n/a)

Added	07 Aug 2010
Updated	07 Aug 2010
Type	Conference
Year	1996
Where	HPDC
Authors	Jarek Nieplocha, Robert J. Harrison

Comments (0)