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

Speedup bounds and processor allocation for parallel programs on multiprocessors

Conference ·
OSTI ID:7163000

The main aim of this paper is to study allocation of processors to parallel programs executing on a multiprocessor system, and the resulting speedups. We first consider a general program represented as a sequence of steps consisting of parallel operations, then one represented as a task graph whose nodes are do across loops and whose edges represent precedence constraints, and finally a single do across loop. General bounds on program speedup are discussed and measurements of code parallelism for the LINPACK numerical package are presented. 11 refs., 4 figs.

Research Organization:
Illinois Univ., Urbana (USA). Center for Supercomputing Research and Development; Control Data Corp., Sunnyvale, CA (USA)
DOE Contract Number:
FG02-85ER25001
OSTI ID:
7163000
Report Number(s):
DOE/ER/25001-9; CONF-860877-3; CSRD-551; ON: DE87002099
Country of Publication:
United States
Language:
English

Similar Records

Processor allocation for horizontal and vertical parallelism and related speedup bounds
Journal Article · Tue Mar 31 23:00:00 EST 1987 · IEEE Trans. Comput.; (United States) · OSTI ID:6626865

Deadlock prevention in processor self-scheduling for parallel nested loops
Conference · Wed Dec 31 23:00:00 EST 1986 · OSTI ID:5768508

Intelligent allocation algorithm for parallel processing
Technical Report · Thu Dec 31 23:00:00 EST 1987 · OSTI ID:5180906