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

Data prefetching in shared memory multiprocessors

Conference ·
OSTI ID:5703538

The trace driven simulation of 16 numerical subroutines is used to compare instruction lookahead and data prefetching with private caches in shared memory multiprocessors with hundreds or thousands of processors and memory modules interconnected with a pipelined network. These multiprocessors are characterized by long memory access delays that create a memory access bottleneck. Using the multiprocessor cache model for comparison, data prefetching is found to be more effective than caches in addressing the memory access bottleneck. 5 refs., 6 figs.

Research Organization:
Illinois Univ., Urbana (USA). Center for Supercomputing Research and Development
DOE Contract Number:
FG02-85ER25001
OSTI ID:
5703538
Report Number(s):
DOE/ER/25001-41; CONF-870831-15; ON: DE88003576
Country of Publication:
United States
Language:
English