Since 1978
Published in Sarov (Arzamas-16), Nizhegorodskaya oblast

RUSSIAN FEDERAL
NUCLEAR CENTER -
ALL-RUSSIAN RESEARCH INSTITUTE
OF EXPERIMENTAL PHYSICS
 
 Русский |  English
ABOUT EDITORIAL BOARD PUBLICATION ETHICS RULES FOR AUTHORS AUTHORS ARCHIVE MOST RECENT ISSUE IN NEXT ISSUE PAPER OF THE YEAR




THE GRAPHICAL PREPROCESSOR SKAT FOR 2D CALCULATIONS

A.V. Kvitchansky, S.N. Lebedev, V.N. Pisarev, O.V. Stryakhnina
VANT. Ser.: Mat. Mod. Fiz. Proc 2000. Вып.4. С. 45-52.

One of the possible approaches to the problem of data preprocessing for the 2D continuum mechanic problem calculations is presented in this paper. SKAT program allows constructing and editing geometrical objects, generating grids according to different methods and describing grid's data.
    The authors aimed to make data presentation process simple, visual and the most fault tolerant. The code has a modern interface typical of windows applications. The visual representation of geometrical data is available with any grain size. The loading operation does not require a strict sequence of operations. Earlier input data can be looked through and edited in graphic and dialog mode. The correctness of user's actions is being tested persistently and if it is necessary the hints and comments are given in the process. Close control over editing operations allows preventing errors completely.



PARALLELIZATION IN COMPUTATIONAL DOMAINS IN SATURN-3 CODES FOR 2D TRANSPORT PROBLEMS

M.E. Lebedeva, S.P. Santalov, S.V. Subbot
VANT. Ser.: Mat. Mod. Fiz. Proc 2000. Вып.4. С. 58-60.

An algorithm is described for parallelization in computational domains for 2D transport problems. The efficiency of this algorithm is studied using computations performed on multi-processors with distributed memory.
    The algorithm is derived from the principle of geometrical decomposition of the original domain in subdomains and performance of separate computations by subdomains; mathematical (computational) domains are chosen for the latter. Computational domains exchange boundary conditions. Iterations by boundary conditions are combined with iterations by right-hand side of the transport equation.
    MPI protocol is chosen for exchanging messages between processors. The use of standard functions of synchronous and asynchronous transmissions allows transferring easily programs on different platforms of translators and operating systems, supporting MPI.










[ Back ]


 
 
 
© FSUE "RFNC-VNIIEF", 2000-2024