Structured asynchronous communication routines for the FPS T-series

Authors:
B. R. Carlile;D. Miles
Affiliations:
Floating Point Systems, Inc., 3601 SW Murray Blvd., Beaverton, Oregon;Floating Point Systems, Inc., 3601 SW Murray Blvd., Beaverton, Oregon
Venue:
C3P Proceedings of the third conference on Hypercube concurrent computers and applications: Architecture, software, computer systems, and general issues - Volume 1
Year:
1988

Citing 6
Cited 2

A VLSI architecture for concurrent data structures

A VLSI architecture for concurrent data structures
Data movement techniques for the pyramid computer

SIAM Journal on Computing
Programming for Parallelism

Computer
Communication effect basic linear algebra computations on hypercube architectures

Journal of Parallel and Distributed Computing
Hypercube algorithms and implementations

SIAM Journal on Scientific and Statistical Computing
The second generation FPS T series: an enhanced parallel vector supercomputer

C3P Proceedings of the third conference on Hypercube concurrent computers and applications: Architecture, software, computer systems, and general issues - Volume 1

The second generation FPS T series: an enhanced parallel vector supercomputer

C3P Proceedings of the third conference on Hypercube concurrent computers and applications: Architecture, software, computer systems, and general issues - Volume 1
A flexible system call interface for interprocessor communication in a distributed memory multicomputer

ACM SIGOPS Operating Systems Review

Quantified Score

Hi-index	0.00

Visualization

Abstract

An innovative programming environment for distributed computing has been developed for the new FPS T Series Parallel Vector Supercomputer. This programming model provides structured asynchronous communication routines, dynamic configuration of processing nodes into application topologies, and external data partitioning and distribution. The asynchronous communication routines have been written to enhance the inter-processor communication capabilities of the T Series. These routines span a wide cross-section of the structured communications necessary for distributed memory scientific computer applications. This includes nearest-neighbor communication, as well as data broadcast, data collection, and data rotation within a logical topology. There is also a routine which enables the user to specify custom communication patterns. Each routine is invoked by a single FORTRAN or C call to perform asynchronous communication in parallel with computation. The implementations of the communication routines use buffering, load balancing, and pipelining techniques. These routines allow one to quickly code high level parallel programs that are efficient as well as readable.