Mapping of micro data flow computations on parallel microarchitectures

Authors:
L. Shih;C. A. Papachristou
Affiliations:
Computer Engineering & Science Department, Case Western Reserve University, Cleveland, Ohio;Computer Engineering & Science Department, Case Western Reserve University, Cleveland, Ohio
Venue:
MICRO 21 Proceedings of the 21st annual workshop on Microprogramming and microarchitecture
Year:
1988

Citing 4
Cited 3

Algorithms, software and hardware of parallel computers

Algorithms, software and hardware of parallel computers
High-performance computer architecture

High-performance computer architecture
Flow graph representation

DAC '86 Proceedings of the 23rd ACM/IEEE Design Automation Conference
Principles Firmware Engineering

Principles Firmware Engineering

A model for microarchitecture structure evaluation

MICRO 22 Proceedings of the 22nd annual workshop on Microprogramming and microarchitecture
Microprogramming heritage of RISC design

MICRO 23 Proceedings of the 23rd annual workshop and symposium on Microprogramming and microarchitecture
Adaptive latency-aware parallel resource mapping: task graph scheduling onto heterogeneous network topology

Proceedings of the Conference on Extreme Science and Engineering Discovery Environment: Gateway to Discovery

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a method for mapping computation algorithms to parallel machines architectures. The approach is based on a fine grain mapping system, FGMS, whose basic rationale is to achieve better matchings between computations and architectures. FGMS consists of four stages, i.e., data flow graph generation, vertical mapping into fine grain graphs, horizontal mapping across interconnected processors and instruction or microcode generation for individual processors.