High-level real-time distributed programming

Abstract

To program real-time distributed applications, programmers have to choose between the general purpose distributed programming languages or a combination of a sequential language and low-level services offered by network-oriented operating systems. However, both solutions prove to be unsatisfactory. While the former alternative rarely meets application needs, the latter usually yields poor software which is difficult to maintain and re-structure. This approach suggests the use of a simple concurrency model well-suited for expressing both real-time and real parallelism. It is based on a language extension providing features for high-level real-time distributed programming. Such a language extension can be easily embedded in any sequential language (e.g. C, C++ or Modula-2). This allows programmers to develop complex systems using their favourite language extended with a small set of new constructs.<>