50 search results - page 5 / 10 » Using SCTP to hide latency in MPI programs |
ICPP
2008
IEEE
15 years 4 months ago
2008
IEEE
An asynchronous work-stealing implementation of dynamic load balance is implemented using Unified Parallel C (UPC) and evaluated using the Unbalanced Tree Search (UTS) benchmark ...
CLUSTER
2006
IEEE
15 years 1 months ago
2006
IEEE
Effective overlap of computation and communication is a well understood technique for latency hiding and can yield significant performance gains for applications on high-end compu...
ASPLOS
1994
ACM
15 years 1 months ago
1994
ACM
There is an increasing trend to use commodity microprocessors as the compute engines in large-scale multiprocessors. However, given that the majority of the microprocessors are so...
PLDI
1995
ACM
15 years 1 months ago
1995
ACM
Traditional list schedulers order instructions based on an optimistic estimate of the load latency imposed by the hardware and therefore cannot respond to variations in memory lat...
SNPD
2003
14 years 11 months ago
2003
This paper focuses on non-strict processing, optimization, and partial evaluation of MPI programs which use incremental data structures (ISs). We describe the design and implement...