Deadlock-Free Message Routing in Multiprocessor Interconnection Networks
IEEE Transactions on Computers
Communication effect basic linear algebra computations on hypercube architectures
Journal of Parallel and Distributed Computing
A parallel triangular solver for distributed-memory multiprocessor
SIAM Journal on Scientific and Statistical Computing
A new method for solving triangular systems on distributed-memory message-passing multiprocessors
SIAM Journal on Scientific and Statistical Computing
Ordered fast Fourier transforms on a massively parallel hypercube multiprocessor
Journal of Parallel and Distributed Computing
Fortran 90 explained
Block-cyclic dense linear algebra
SIAM Journal on Scientific Computing
Massively Parallel Linpack Benchmark on the Intel Touchstone Delta andIPSC/860 Systems (Progress Report)
Combinatorial Algorithms: Theory and Practice
Combinatorial Algorithms: Theory and Practice
HPFBench: a high performance Fortran benchmark suite
ACM Transactions on Mathematical Software (TOMS)
Compiling stencils in high performance Fortran
SC '97 Proceedings of the 1997 ACM/IEEE conference on Supercomputing
| Hi-index | 0.00 |
We describe the use and implementation of a polyshift functionPSHIFT for circular shifts and end-offs shifts. Polyshift is usefulin many scientific codes using regular grids, such as finitedifference codes in several dimensions, and multigrid codes,molecular dynamics computations, and in lattice gauge physicscomputations, such as quantum chromodynamics (QCD) calculations.Our implementation of the PSHIFT function on the Connection Machinesystems CM-2 and CM-200 offers a speedup of up to a factor of 3-4compared with CSHIFT when the local data motion within a node issmall. The PSHIFT routine is included in the Connection MachineScientific Software Library (CMSSL).