Accession Number : ADA308175

Title :   Multithreaded Model for Dynamic Load Balancing Parallel Adaptive PDE Computations.

Descriptive Note : Contractor rept.,

Corporate Author : INSTITUTE FOR COMPUTER APPLICATIONS IN SCIENCE AND ENGINEERING HAMPTON VA

Personal Author(s) : Chrisochoides, Nikos

PDF Url : ADA308175

Report Date : NOV 1995

Pagination or Media Count : 27

Abstract : We present a multithreaded model for the dynamic load-balancing of numerical, adaptive computations required for the solution of Partial Differential Equations (PDEs) on multiprocessors. Multithreading is used as a means of exploring concurrency in the processor level in order to tolerate synchronization costs inherent to traditional (non-threaded) parallel adaptive PDE solvers. Our preliminary analysis for parallel, adaptive PDE solvers indicates that multithreading can be used as a mechanism to mask overheads required for the dynamic balancing of processor workloads with computations required for the actual numerical solution of the PDEs. Also, multithreading can simplify the implementation of dynamic load-balancing algorithms, a task that is very difficult for traditional data parallel adaptive PDE computations. Unfortunately, multithreading does not always simplify program complexity, often makes code re-usability not an easy task, and increases software complexity.

Descriptors :   *DISTRIBUTED DATA PROCESSING, *COMPUTER COMMUNICATIONS, *PARALLEL PROCESSING, MATHEMATICAL MODELS, ALGORITHMS, SOFTWARE ENGINEERING, COMPUTATIONS, DATA MANAGEMENT, NUMERICAL ANALYSIS, INPUT OUTPUT PROCESSING, CONCURRENT ENGINEERING, MATHEMATICAL PROGRAMMING, FORTRAN, ADAPTIVE SYSTEMS, PARTIAL DIFFERENTIAL EQUATIONS, OPERATING SYSTEMS(COMPUTERS), MULTIPROCESSORS, SYSTEMS ANALYSIS, COMPUTER NETWORKS, WORKLOAD, MESSAGE PROCESSING.

Subject Categories : Computer Programming and Software
      Computer Systems

Distribution Statement : APPROVED FOR PUBLIC RELEASE