Abstract
Stream processing is a well-suited application pattern for embedded computing. This holds true even more so when it comes to multi-core systems where concurrency plays an important role. With the latest trend towards more dynamic and heterogeneous systems there seems to be a shift from purely synchronous systems towards more asynchronous
ones. The downside of this shift is an increase in programming
complexity due to the more subtile concurrency issues. Several special purpose streaming languages have been proposed to help the programmer in coping with these concurrency issues. In this paper, we take a different approach. Rather than proposing a full-blown programming language, we propose a coordination language named PS-Net. Its purpose is to coordinate existing resource-bound building blocks by means of asynchronous streaming. Within this paper we introduce code annotations and synchronisation patterns that result in a flexible but still
resource-boundable coordination language At the example of a
raytracing application we demonstrate the applicability of PS-Net for expressing the coordination of rather dynamic computations in a resource-bound way.
ones. The downside of this shift is an increase in programming
complexity due to the more subtile concurrency issues. Several special purpose streaming languages have been proposed to help the programmer in coping with these concurrency issues. In this paper, we take a different approach. Rather than proposing a full-blown programming language, we propose a coordination language named PS-Net. Its purpose is to coordinate existing resource-bound building blocks by means of asynchronous streaming. Within this paper we introduce code annotations and synchronisation patterns that result in a flexible but still
resource-boundable coordination language At the example of a
raytracing application we demonstrate the applicability of PS-Net for expressing the coordination of rather dynamic computations in a resource-bound way.
| Original language | English |
|---|---|
| Title of host publication | In: Proceedings of the 1st International Conference on Computational Logics, Algebras, Programming, Tools, and Benchmarking (COMPTOOL'10) |
| Publisher | International Academy, Research and Industry Association |
| Pages | 22-28 |
| ISBN (Print) | 978-1-61208-001-7 |
| Publication status | Published - 2010 |
Keywords
- stream processing
- embedded systems
- multi-core
- resource-constrained