Optimizing horizontal microprograms for vectorial loops with timed petri nets

Authors:
C. Hanen
Affiliations:
Univ. of Paris, France
Venue:
ICS '88 Proceedings of the 2nd international conference on Supercomputing
Year:
1988

Citing 5
Cited 0

Bulldog: a compiler for VLSI architectures

Bulldog: a compiler for VLSI architectures
Timed Petri net schedules

Advances in Petri Nets 1988
Local Microcode Compaction Techniques

ACM Computing Surveys (CSUR)
Some scheduling techniques and an easily schedulable horizontal architecture for high performance scientific computing

MICRO 14 Proceedings of the 14th annual workshop on Microprogramming
Improving the throughput of a pipeline by insertion of delays

ISCA '76 Proceedings of the 3rd annual symposium on Computer architecture

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper deals with the minimization of the asymptotic latency of horizontal microprograms performing vectorial loops. On a quite general architecture model, we show that all the constraints can be formalized to give a cyclic scheduling problem, modelled with a Timed Petri net. To analyse the net behaviour, a bivalued resolution graph is build. A critical cycle of this graph leads to optimal microprograms whose structure is independent on the number of iterations. This approach is then extended to uniform recurrences.