ПРЕДСТАВЛЕНИЕ ГРАФОВ С АССОЦИАТИВНЫМИ ОПЕРАЦИЯМИ НА ЯЗЫКЕ ПРОГРАММИРОВАНИЯ SET@L

Abstract

Usually, an information graph with associative operations has a sequential (“head/tail”) orparallel (“half-splitting”) topology with invariable quantity of operational vertices. If computationalresource is insufficient for the implementation of all vertices, the reduction transformationsof graphs with basic topologies do not allow for the creation of an efficient resource-independentprogram. In fact, the “half-splitting” variant is characterized by irregular connections betweeniterations, and the “head/tail” structure has an increased data duty cycle in the reduced form.In this paper, we propose to transform the topology of a graph with associative operations into acombined variant with sequential and parallel fragments of calculations. The resultant combinedtopology depends on computational resource of a parallel computer system, and such transformationprovides the improvement of specific performance for the reduced computing structure.The considered topology contains isomorphic subgraphs with the “half-splitting” topology, whichinclude the maximal number of hardwarily implemented operational vertices, but the processing ofintermediate data is performed using the “head/tail” principle. The computing structure for thecombined topology has minimal latency and includes one basic subgraph and one vertex withfeedback. This vertex is obtained as a result of the “head/tail” block reduction. We develop analgorithm for the conversion of the initial sequential graph to various combined topologies or tothe limiting case of the “half-splitting” topology with regard to available hardware resource.Within traditional methods of parallel programming, it is possible to describe the variety of topologiesonly as a set of separated subprograms. To create an efficient resource-independent program,we propose the application of the Set@l programming language. We describe the“head/tail” and “half-splitting” principles as the attributes of set processing methods in Set@l.Resource-independent program uses these types and parallelism attributes for the modification oftopology and further reduction of performance in the corresponding aspects.