SLICC: A low latency interface for collective communications
Conference
·
OSTI ID:87614
- Purdue Univ., West Lafayette, IN (United States)
- Cray Research Inc., Eagan, MN (United States)
Several recent parallel computers have implemented logically shared, physically distributed memory systems which allow processors to directly access memory in other processors without interrupting the referenced PE. Because this kind of architecture provides greater flexibility for interprocessor communications than private address space computers, different software models can be developed to take advantage of these machines. In this paper, the authors describe a low-level collective communications interface specifically tailored for shared address space parallel computers. They also provide performance results for an implementation on the Cray T3D and compare its performance to Cray`s PVM implementation. These results show that SLICC can be more than 10 times faster than PVM on the T3D.
- OSTI ID:
- 87614
- Report Number(s):
- CONF-941118--; ISBN 0-8186-6605-6
- Country of Publication:
- United States
- Language:
- English
Similar Records
Implementing a parabolized Navier-Stokes flow solver on the Cray T3D
Parallel algorithms for modeling flow in permeable media. Annual report, February 15, 1995 - February 14, 1996
DANTSYS/MPI: a system for 3-D deterministic transport on parallel architectures
Conference
·
Thu Nov 30 23:00:00 EST 1995
·
OSTI ID:125485
Parallel algorithms for modeling flow in permeable media. Annual report, February 15, 1995 - February 14, 1996
Technical Report
·
Thu Mar 14 23:00:00 EST 1996
·
OSTI ID:764425
DANTSYS/MPI: a system for 3-D deterministic transport on parallel architectures
Conference
·
Mon Dec 30 23:00:00 EST 1996
·
OSTI ID:459820