PCT 04/2016

Parallel computational technologies (PCT) 2016 was held on March 28th - April 1st, 2016 in Russia, Arkhangelsk, Northern (Arctic) Federal University.

A.A. Smirnov, N.A. Kataev have proposed there a speaker paper:

About approaches to program transformation in SAPFOR system

The result of applying a fixed, program-independent sequence of optimizations when compiling programs doesn't reflect the features of their information structure. It affects the performance of parallel programs obtained by use of  the non-adaptive automatically parallelizing compilers. The iterative process of program parallelizing, supported by  SAPFOR system of automated parallelization, allows  to select the needed transformations and apply them only when they are necessary to eliminate the problems preventing parallelization.  

This article is written by a team of the following authors M.S. Baranov, N.A. Kataev, A.A. Smirnov.

A.S. Kolganov, M.N. Pritula, V.A. Bakhtin have proposed there a speaker paper:

DVMH model extension for operation with irregular grids

DVMH model is suitable first of all for writing of parallel programs on the regular rectangular grids, but it is possible to parallelize some types of programs on the irregular grids using available tools. There is offered a variant of DVMH model extension which, on the one hand, would organically be inscribed in the existing DVMH model, adding its constructions, and on the other hand would allow to remove the known problems and restrictions in the task parallelization on irregular grids, and not losing considerably in efficiency of parallel execution.  

This article is written by a team of the following authors V.A. Bakhtin, A.S. Kolganov, V.A. Krukov, N.V. Podderugina, S.V. Polyakov, M.N. Pritula.

A.S. Kolganov has proposed there a speaker paper:

Parallel implementation of the search algorithm for minimum spanning trees using central and graphic processors

The solution of the problem to search a minimum spanning trees is widespread in various fields of researches: recognition of different objects, computer vision,  analysis and creation of networks (for example, telephone, electrical, computer, road, etc.), chemistry and biology and many others. There are at least three well-known algorithms that solve this problem: Boruvka's, Kruskal's and Prim's.  Large graph processing is a time-consuming task for Central processor unit (CPU) and is popular at present. The graphic accelerators (GPU) having greater computational power, than CPU, become more and more widespread for solution of  general purpose tasks.  But this problem, as well as many graph processing tasks, badly lays down on GPU architecture. In this article hybrid implementation of the algorithm will be considered.  

This article is written by A.S. Kolganov.