Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Accelerating Numerical Calculation on the Cray XMT

Conference ·
OSTI ID:974008
The Cray XMT provides hardware support for parallel algorithms that would be communication- or memory-bound on other machines. Unfortunately, even if an algorithm meets these criteria, performance suffers if the algorithm is too numerically intensive. We present a lookup-based approach that achieves a significant performance advantage over explicit calculation. We describe an approach to balancing memory bandwidth against on-chip floating point capabilities, leading to further speedup. Finally, we provide table lookup algorithms for a number of common functions.
Research Organization:
Pacific Northwest National Laboratory (PNNL), Richland, WA (US)
Sponsoring Organization:
USDOE
DOE Contract Number:
AC05-76RL01830
OSTI ID:
974008
Report Number(s):
PNNL-SA-64423; 400470000
Country of Publication:
United States
Language:
English