P. Lopez et al., A METHODOLOGY FOR GRANULARITY-BASED CONTROL OF PARALLELISM IN LOGIC PROGRAMS, Journal of symbolic computation, 21(4-6), 1996, pp. 715-734
Citations number
26
Categorie Soggetti
Mathematics,"Computer Sciences, Special Topics",Mathematics,"Computer Science Theory & Methods
Several types of parallelism can be exploited in logic programs while
preserving correctness and efficiency, i.e. ensuring that the parallel
execution obtains the same results as the sequential one and the amou
nt of work performed is not greater. However, such results do not take
into account a number of overheads which appear in practice, such as
process creation and scheduling, which can induce a slow-down, or, at
least, limit speedup, if they are not controlled in some way. This pap
er describes a methodology whereby the granularity of parallel tasks,
i.e. the work available under them, is efficiently estimated and used
to limit parallelism so that the effect of such overheads is controlle
d. The run-time overhead associated with the approach is usually quite
small, since as much work is done at compile time as possible. Also,
a number of run-time optimizations are proposed. Moreover, a static an
alysis of the overhead associated with the granularity control process
is performed in order to decide its convenience. The performance impr
ovements resulting from the incorporation of grain size control are sh
own to be quite good, specially for systems with medium to large paral
lel execution overheads. (C) 1996 Academic Press Limited