A proposal for a heterogeneous cluster ScaLAPACK (dense linear solvers)

IEEE Transactions on Computers

Volumn 50, Issue 10, 2001, Pages 1052-1070

  Beaumont, Olivier ^b   Boudet, Vincent ^b   Petitet, Antoine ^b   Rastello, Fabrice ^b   Robert, Yves ^a,b  

^a IEEE   (France)

^b UNIVERSITÉ DE LYON   (France)

Author keywords

Cluster computing; Data allocation; Data distribution; Different speed processors; Heterogeneous grid; Heterogeneous network; Heterogeneous platforms; Load balancing; Numerical libraries; Numerical linear algebra

Indexed keywords

HETEROGENEOUS NETWORKS;

LINEAR ALGEBRA; LINEAR SYSTEMS; PARALLEL PROCESSING SYSTEMS; PROGRAM PROCESSORS;

COMPUTER WORKSTATIONS;

EID: 0035481895     PISSN: 00189340     EISSN: None     Source Type: Journal    
DOI: 10.1109/12.956091     Document Type: Article

References (37)

1. A high performance matrix multiplication algorithm on a distributed-memory parallel computer, using overlapped communication
2. Parallel application scheduling on networks of workstations
3. A parallel language and its programming system for heterogeneous networks
4. Partitioning a square into rectangles: Np-completeness and approximation algorithms
5. High-performance schedulers
6. Scalapack: A portable linear algebra library for distributed-memory computers-design issues and performance
7. Static tiling for heterogeneous computing platforms
8. ScaLAPACK: A portable linear algebra library for distributed memory computers-design issues and performance
9. The design and implementation of the ScaLAPACK LU, QR, and cholesky factorization routines
10. QR factorization of a dense matrix on a hypercube multiprocessor
11. Scheduling algorithms for heterogeneous network of workstations
12. The limited applicability of block decomposition in cluster computing
13. Block data decomposition for data-parallel programming on a heterogeneous workstation network
14. Scalability issues in the design of a library for dense linear algebra
15. Software libraries for linear algebra computations on high performance computers
16. Matrix algorithms on a hypercube I: Matrix multiplication
17. Approximation algorithm for partitioning a rectangle with interior points
18. On the complexity of the generalized block distribution
19. Dynamic, competitive scheduling of multiple dags in a distributed heterogeneous environment
20. Array decompositions for nonuniform computational environments
21. Heterogeneous distribution of computations while solving linear algebra problems on networks of heterogeneous computers
22. On approximating rectangle tiling and packing
23. Minimum edge length partitioning of rectilinear polygons
24. A dynamic matching and scheduling algorithm for heterogeneous computing systems
25. Software support for heterogeneous computing
26. A compile-time scheduling heuristic for interconnection-constrained heterogeneous processor architectures
27. Minimizing the application execution time through scheduling of subtasks and communication traffic in a heterogeneous computing system
28. Scheduling parallel applications in distributed networks
29. Automatically tuned linear algebra software