Efficient Compilation of Esterel for Multi-core Execution

Recent advances in processor technology have lead to affordable multi-core processors, which could even be used in embedded applications. However, many embedded applications are safety-critical and require suitable abstractions such as the synchronous abstraction. Esterel is one such language belonging to the synchronous family and has been used extensively in the design of safety critical systems. While several compilation techniques of Esterel have been proposed, these are unsuitable for multi-cores due to the inherently sequential approach of \emph{compiling away} the concurrency. We overcome this limitation by proposing two distinct approaches that distribute Esterel threads evenly across multi-core architectures. The first approach statically distributes threads based on the computation intensity approximated by the number of instructions generated from each thread. The second approach distributes threads dynamically using a thread queue that dispatches a thread whenever a core becomes idle. We have performed extensive benchmarking over large Esterel programs to illustrate that, using the static approach, achieving throughput with parallel execution of Esterel is benchmark dependent. However, the dynamic approach not only benefits data-dominated Esterel programs, but also large control-dominated ones. In particular, gains in performance of 36% and 93% were attained for a large control-dominated program using a dual-core and quad-core Microblaze processor, respectively.

Data and Resources

Additional Info

Field Value
Source https://inria.hal.science/hal-00728149
Author Yuan, Simon, Yoong, Li Hsien, Roop, Partha S.
Maintainer CCSD
Last Updated November 3, 2026, 21:28 (UTC)
Created November 3, 2026, 21:28 (UTC)
Identifier Report N°: RR-8056
Language en
Rights https://about.hal.science/hal-authorisation-v1/
contributor Department of Electrical and Computer Engineering [Auckland ] (ECE) ; University of Auckland [Auckland]
creator Yuan, Simon
date 2012-09-05T00:00:00
harvest_object_id 38dd237f-da66-483a-9736-c75c10d2c0af
harvest_source_id 3374d638-d20b-4672-ba96-a23232d55657
harvest_source_title test moissonnage SELUNE
metadata_modified 2026-10-23T00:00:00
set_spec type:REPORT