Algorithmic modifications to the Jacobi-Davidson parallel eigensolver to dynamically balance external CPU and memory load

Proceedings of the International Conference on Supercomputing

Volumn , Issue , 2001, Pages 454-463

  Mills, R I ^a   Stathopoulos, A ^a   Smirni, E ^a  

^a The College of William and Mary   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

COMPUTER WORKSTATIONS; CONVERGENCE OF NUMERICAL METHODS; DIGITAL STORAGE; EIGENVALUES AND EIGENFUNCTIONS; ITERATIVE METHODS; PARALLEL ALGORITHMS; PROBLEM SOLVING;

MEMORY LOADS;

PARALLEL PROCESSING SYSTEMS;

EID: 0034817703     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1145/377792.377903     Document Type: Conference Paper

References (31)

1. Scheduling from the perspective of the application networks
2. Application level scheduling on distributed heterogeneous networks
3. A portable programming interface for performance evaluation on modern processors
4. A scalable cross-platform infrastructure for application performance tuning using hardware counters
5. Complexity of memory-efficient kronecker operations with applications to the solution markov models
6. Automatic configuration and run-time adaptation of distributed applications
7. Jacobi-Davidson style QR and QZ algorithms for the partial reduction of matrix pencils
8. Globus: A metacomputing infrastructure toolkit
9. The Legion vision of a worldwide virtual computer
10. Multilevel diffusion schemes for repartitioning of adaptive meshes
11. A parallel algorithm for multilevel graph partitioning and sparse matrix ordering
12. Exploiting application alternatives
13. Non-standard parallel solution strategies for distributed sparse linear systems
14. ILUT: A dual threshold incomplete ILU factorization
15. A parallel, block, Jacobi-Davidson implementation for solving large eigenproblems on coarse grain environments
16. Parallel, multi-grain eigensolvers with applications to materials science
17. Parallel methods and tools for predicting material properties
18. The network weather service: A distributed resource performance forecasting service for metacomputing
19. Predicting the CPU availability of time-shared unix systems