Speed-up of scalable iterative linear solvers implemented on an array of transputers

Parallel Computing

Volumn 20, Issue 3, 1994, Pages 375-387

  Asenov, A ^a   Reid, D ^a   Barker, J R ^a  

^a UNIVERSITY OF GLASGOW   (United Kingdom)

Author keywords

Parallel iterative solvers; Processor array; Speedup prediction; Transputers

Indexed keywords

ALGORITHMS; ARRAYS; CODES (SYMBOLS); COMPUTATION THEORY; COMPUTATIONAL COMPLEXITY; COMPUTER SIMULATION; DIFFERENTIAL EQUATIONS; FORTRAN (PROGRAMMING LANGUAGE); ITERATIVE METHODS; LARGE SCALE SYSTEMS; PROGRAM PROCESSORS; RESPONSE TIME (COMPUTER SYSTEMS);

CALCULATION TIME; LOCAL COMMUNICATION; OSCILLATORY BEHAVIOR; PARALLEL LINEAR ITERATIVE SOLVERS; PARSYTEC SUPER CLUSTER MODEL; PROBLEM SIZES; PROCESSOR ARRAYS; RECTANGULAR DISCRETE GRID; SPEEDUP PREDICTION; TRANSPUTERS;

PARALLEL PROCESSING SYSTEMS;

EID: 0028392846     PISSN: 01678191     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0167-8191(06)80020-7     Document Type: Article

References (20)

1. Towards developing robust algorithms for solving partial differential equations on MIMD machines
2. Scheduling parallel iterative methods on multiprocessor systems
3. Parallel S.O.R. iterative methods
4. A parallel point row-wise successive over-relaxation method on a multiprocessor
5. Two parallel SOR variants of the Schwarz alternating procedure
6. Block colouring schemes for the SOR method on local memory parallel computers
7. SOR and MGR[v] experiments on Crystal multicomputer
8. Successive underrelaxation (SUR) and generalised conjugate gradient (GCG) methods for hyperbolic difference equations on a parallel computer
9. MIMD implementations of linear solvers for oil reservoir simulation
10. A parallel multigrid FAS scheme for transputer networks
11. Efficiency of some parallel numerical algorithms on distributed systems
12. Performance of parallel processors
13. Performance modelling of distributed memory architectures
14. A methodology for performance evaluation of parallel applications on multiprocessors
15. Straightforward partitioning of composite grids for explicit difference methods
16. Communications overhead and the expected speed-up of multidimensional mesh-connected parallel processors
17. A multicolor SOR method for parallel computation
18. The implementation of the explicit block iterative method on the balance 8000 parallel computer