SYRF Project

Task 4/5.1: "Integrating synchrony & asynchrony: fundamentals"

Abstract of deliverable 4/5.1.2

Back to the SYRF Home Page



Foreword:

Integrating synchrony & asynchrony is a major issue for the practical use of the synchronous technology, this is why a WP is devoted to it. This WP and the present task summary is presented according to the new organization for the former two WP 4 and 5. Accordingly, a task called fundamentals was introduced in this reorganization, and we report its progresses now.

Approach:

There are several issues that need to be considered in fundamental studies related to synchrony & asynchrony, namely:

  1. How to distribute correctly a synchronous program on a given, possibly asynchronous, architecture? What are the available theoretical results? What are the assumptions required, for the supporting execution architecture?
  2. In what extend a property proved for the original synchronous program is still valid for the same one, running on a distributed, possibly asynchronous, architecture? 
  3. What support can be provided (with theoretical justifications) for the reuse of pre-existing components? How can encapsulation be guaranteed correct?

This is the type of question which we want to provide theoretical support for. The difficulties we are faced with are the following:

Achieved Results:

Documents: we have prepared a deliverable focussing on the results regarding desynchronisation; it is basically extracted from reference 2 below. This reference also includes additional material, related to causality analysis, reasoning on scheduling, with separate compilation as an objective.

Publications:

  1. A. Benveniste, P. Le Guernic, and P. Aubry
    Compositionality in dataflow synchronous languages: specification and code generation,
    Malente Workshop on Compositionality, W.P. de Roever, A. Pnueli Eds., September 1997, to appear in LNCS, late 1998 or early 1999.
  2. A. Benveniste, P. Le Guernic, B. Caillaud, and P. Aubry
    Compositionality in dataflow synchronous languages: specification and code generation,
    (extended version of the former with a detailed study of desynchronisation),
    submitted to I&C.
  3. A. Benveniste
    Safety Critical Embedded Systems Design: the SACRES approach,
    3 hour course given at FTRTFT'98 school on the whole SACRES project,
    manuscript available.