The Lanczos algorithm for the generalized symmetric eigenproblem on shared-memory architectures

Conference · Sat Sep 01 04:00:00 EDT 1990

OSTI ID:5748587

Jones, M T; Patrick, M L

The generalized eigenvalue problem, Kx = {lambda}Mx, is of significant practical importance, for example, in structural engineering where it arises as the vibration and buckling problems. The paper describes the implementation of a solver based on the Lanczos algorithm, LANZ, on two shared-memory architectures, the CRAY Y-MP and Encore Multimax. Issues arising from implementing linear algebra operations on a multivector processor are examined. Portability between a multivector processor and a simple multiprocessor is discussed. A model is developed and used to predict the performance of LANZ on shared-memory architectures. Performance results from some practical problems are given and analyzed. 20 refs., 5 figs.

• Research Organization: Argonne National Lab., IL (USA)
• Sponsoring Organization: DOE; NASA; USDOE, Washington, DC (USA); National Aeronautics and Space Administration, Washington, DC (USA)
• DOE Contract Number: W-31109-ENG-38
• OSTI ID: 5748587
• Report Number(s): CONF-910771-1; ON: DE91006044; CNN: NAS1-18107; NAS1-18605; NAG-1-466
• Country of Publication: United States
• Language: English

Related Subjects

42 ENGINEERING
420200 -- Engineering-- Facilities
Equipment
& Techniques
99 GENERAL AND MISCELLANEOUS
990200* -- Mathematics & Computers
ALGORITHMS
ARRAY PROCESSORS
COMPUTER CALCULATIONS
COMPUTERS
CRAY COMPUTERS
EIGENVALUES
FACTORIZATION
MATHEMATICAL LOGIC
MECHANICAL STRUCTURES
PARALLEL PROCESSING
PERFORMANCE
PROGRAMMING