Network performance-aware collective communication for clustered wide-area systems

Parallel Computing

Volumn 27, Issue 11, 2001, Pages 1431-1456

  Kielmann, Thilo ^a   Bal, Henri E ^a   Gorlatch, Sergei ^b   Verstoep, Kees ^a   Hofman, Rutger F H ^a  

^a VU UNIVERSITY AMSTERDAM   (Netherlands)

^b TECHNISCHE UNIVERSITÄT BERLIN   (Germany)

Author keywords

Collective communication for clustered wide area systems; Grid computing; MPI

Indexed keywords

BANDWIDTH; HIERARCHICAL SYSTEMS; MULTITASKING; OPTIMIZATION; RESPONSE TIME (COMPUTER SYSTEMS); WIDE AREA NETWORKS;

CLUSTERED WIDE-AREA SYSTEMS; COLLECTIVE COMMUNICATIONS;

PARALLEL PROCESSING SYSTEMS;

EID: 0035480270     PISSN: 01678191     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0167-8191(01)00098-9     Document Type: Article

References (38)

1. Log GP: Incorporating long messages into the Log P model - One step closer towards a realistic model for parallel computation
2. Performance evaluation of the Orca shared object system
3. Efficient collective communication on heterogeneous networks of workstations
4. Collective communication operations: Experimental results vs. theory
5. User-level network interface protocols
6. Myrinet: A gigabit-per-second local area network
7. On the design and implementation of broadcast and global combine operations using the postal model
8. Metacomputing
9. Log P: Towards a realistic model of parallel computation
10. Assessing fast network interfaces
11. MPI_Connect: Managing heterogeneous MPI applications interoperation and process control
12. Globus: A metacomputing infrastructure toolkit
13. Distributed computing in a heterogeneous computing environment
14. Status report on the development of the interoperable MPI protocol
15. Algorithms for collective communication operations on SMP clusters
16. Optimization rules for programming with collective operations
17. An enabling framework for master-worker applications on the computatinal grid
18. The legion vision of a worldwide virtual computer
19. A high-performance, portable implementation of the MPI message passing interface standard
20. Improving the performance of MPI derived datatypes
21. MPI-StarT: Delivering network performance to numerical applications
22. Cross-platform analysis of fast messages for Myrinet
23. Exploiting hierarchy in parallel computer networks to optimize collective operation performance
24. Optimal broadcast and summation in the Log P model
25. Optimal multicast with packetization and network interface support
26. Bandwidth-efficient collective communication for clustered wide area systems
27. Fast measurement of Log P parameters for message passing platforms
28. MPI's reduction operations in clustered wide area systems
29. MagPIe: MPI's collective communication operations for clustered wide area systems
30. ECO: Efficient collective operations for communication on heterogeneous networks
31. On efficiently implementing global time for performance evaluation on multiprocessor systems
32. Construction of optimal multicast trees based on the parameterized communication model
33. On calibrating measurements of packet transit times
34. Optimal and near-optimal algorithms for k-item broadcast
35. On global combine operations
36. A pipelined broadcast for multidimensional meshes
37. Forecasting network performance to support dynamic scheduling using the network weather service