Sciweavers

365 search results - page 2 / 73
» Automatic Performance Debugging of SPMD Parallel Programs
Sort
View
PPOPP
2011
ACM
12 years 7 months ago
ScalaExtrap: trace-based communication extrapolation for spmd programs
Performance modeling for scientific applications is important for assessing potential application performance and systems procurement in high-performance computing (HPC). Recent ...
Xing Wu, Frank Mueller
ICCS
2009
Springer
13 years 11 months ago
Hiding Communication Latency with Non-SPMD, Graph-Based Execution
Reformulating an algorithm to mask communication delays is crucial in maintaining scalability, but traditional solutions embed the overlap strategy into the application. We present...
Jacob Sorensen, Scott B. Baden
ICPP
1991
IEEE
13 years 8 months ago
Automatic Parallel Program Generation and Optimization from Data Decompositions
Data decomposition is probably the most successful method for generating parallel programs. In this paper a general framework is described for the automatic generation of parallel...
Edwin M. R. M. Paalvast, Henk J. Sips, Arjan J. C....
IPPS
2007
IEEE
13 years 11 months ago
Formal Analysis for Debugging and Performance Optimization of MPI
High-end computing is universally recognized to be a strategic tool for leadership in science and technology. A significant portion of high-end computing is conducted on clusters...
Ganesh Gopalakrishnan, Robert M. Kirby
ICPP
2007
IEEE
13 years 11 months ago
Toward Automatic Data Distribution for Migrating Computations
Program parallelization requires mapping computation and data to processing elements. Navigational Programming (NavP), based on the principle of migrating computations, offers a d...
Lei Pan, Jingling Xue, Ming Kin Lai, Michael B. Di...