Linear algebra routines can be parallelized and run on a grid using the theory of parallel algorithms.
http://www.cs.utexas.edu/users/plapack/papers/ipps98/ipps98.html
This is Cannons algorithm for matrix multiplication.
http://orca.st.usm.edu/~seyfarth/sc730/cannon.html
MonteCarlo computation is also a good candidate for parallel processing.
The following Java programs (for computational physics) are not parallelised but are interesting nonetheless.
http://www.physics.unlv.edu/~pang/cp2_j.html
Monday, 29 December 2008
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment