Computational models and task scheduling for parallel sparse Cholesky factorization. Final report, 1 October 1986-30 September 1987

Technical Report · Wed Oct 01 00:00:00 EDT 1986

OSTI ID:5384897

Liu, J W

In this paper, a systematic and unified treatment of computational task models for parallel sparse Cholesky factorization is presented. They are classified as fine-, medium-, and large-gained graph models. In particular, a new medium-grained model based on column-oriented tasks is introduced, and it is shown to correspond structurally to the filled graph of the given sparse matrix. The task-scheduling problem for the various task graphs is also discussed. A practical algorithm to schedule the column tasks of the medium-grained model for multiple processors is described. It is based on a heuristic critical path scheduling method. This will give an overall scheme for parallel sparse Cholesky factorization, appropriate for parallel machines with shared-memory architecture like the Denelcor HEP.

• Research Organization: Oak Ridge National Lab., TN (USA)
• OSTI ID: 5384897
• Report Number(s): AD-A-187038/5/XAB
• Resource Relation: Other Information: Pub. in Parallel Computing, Vol. 3, No. 4, 327-342(Oct 1986)
• Country of Publication: United States
• Language: English

Related Subjects

99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
COMPUTER ARCHITECTURE
COMPUTER GRAPHICS
PARALLEL PROCESSING
TASK SCHEDULING
ALGORITHMS
ARRAY PROCESSORS
GRAPHS
MATHEMATICAL MODELS
PROGRESS REPORT
DATA PROCESSING
DOCUMENT TYPES
MATHEMATICAL LOGIC
PROCESSING
PROGRAMMING
990210* - Supercomputers- (1987-1989)