Visual Parallel Programming as PaaS Cloud Service with Graph-Symbolic Programming Technology
Most computer programs are created in textual form. From high-level programming languages to CPU instructions both programmer and computer work with sequences of characters and words. Textual representation of the program combines centuries-old tradition of writing as the universal form of fixing human thoughts with ease of interpretation and analysis of text by computer. The sequential nature of text makes it suitable for description of instruction sequences and sequential algorithms. At the same time the text is inconvenient for clear representation of parallel programs. In such programs it is important to depict instructions that can be executed concurrently. In this case the graphical (visual) representation is more suitable.
In this paper we present the visual approach to parallel programming provided by Graph-Symbolic Programming Technology. This technology uses text to represent small sequential subprograms (mathematical expressions or small methods). Visual representation in graph form is used to depict program logic and concurrency. The basics of this technology are considered as well as advantages and disadvantages of visual parallel programming. Synchronization primitives used in Graph-Symbolic Programming Technology and their visual form are described. The method is proposed for compact and clear representation of multiple similar parallel processes.
The technology is being implemented as a PaaS cloud service that provides the tools for creation, validation and execution of parallel programs on cluster systems. The current state of this work is also presented. We argue that visual programming and cloud technologies provide the capability of shared development of programs and algorithms that text programming lacks. The visual programming in such implementation gains the features of the visual modeling.
Proceedings of the Institute for System Programming, vol. 27, issue 3, 2015, pp. 47-56
ISSN 2220-6426 (Online), ISSN 2079-8156 (Print).
DOI: 10.15514/ISPRAS-2015-27(3)-3Full text of the paper in pdf Back to the contents of the volume