Geodesic
Design of a systolic array for computational solving of a nonstationary equation of heat conductivity
Trudy Instituta matematiki, Tome 15 (2007) no. 2, pp. 104-110.

Voir la notice de l'article provenant de la source Math-Net.Ru

A systolic array of a ring architecture that consists of a given number $\Delta$ of homogeneous processor elements is designed. The array is destined to numerous solution of a nonstationari equation of heat conductivity by explicit net method. The local memory of processor elements does not depend on the parameters $N$ and $M$ that determine the number of mesh points, nor the number of processors $\Delta$. Time of solving the problem is determined by the function $\displaystyle\frac{M(N-3)}{\Delta}+\Delta+2M+1$ that has the minimum value for $\Delta=\sqrt{M(N-3)}$ (under the assumption that $\sqrt{M(N-3)}$ is an integer). 
@article{TIMB_2007_15_2_a10,
     author = {P. I. Sobolevskii},
     title = {Design of a systolic array for computational solving of a nonstationary equation of heat conductivity},
     journal = {Trudy Instituta matematiki},
     pages = {104--110},
     publisher = {mathdoc},
     volume = {15},
     number = {2},
     year = {2007},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/TIMB_2007_15_2_a10/}
}
TY  - JOUR
AU  - P. I. Sobolevskii
TI  - Design of a systolic array for computational solving of a nonstationary equation of heat conductivity
JO  - Trudy Instituta matematiki
PY  - 2007
SP  - 104
EP  - 110
VL  - 15
IS  - 2
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/TIMB_2007_15_2_a10/
LA  - ru
ID  - TIMB_2007_15_2_a10
ER  - 
%0 Journal Article
%A P. I. Sobolevskii
%T Design of a systolic array for computational solving of a nonstationary equation of heat conductivity
%J Trudy Instituta matematiki
%D 2007
%P 104-110
%V 15
%N 2
%I mathdoc
%U http://geodesic.mathdoc.fr/item/TIMB_2007_15_2_a10/
%G ru
%F TIMB_2007_15_2_a10
P. I. Sobolevskii. Design of a systolic array for computational solving of a nonstationary equation of heat conductivity. Trudy Instituta matematiki, Tome 15 (2007) no. 2, pp. 104-110. http://geodesic.mathdoc.fr/item/TIMB_2007_15_2_a10/

[1] Kun S., Matrichnye protsessory na SBIS, per. s angl., Mir, M., 1991

[2] Kung H.T., “Why systolic architecture?”, Computer, 15:1 (1982), 37–46 | DOI

[3] Voevodin V.V., Voevodin Vl.V., Parallelnye vychisleniya, BKhV-Peterburg, SPb., 2002

[4] Likhoded N.A., Sobolevskii P.I., Tiunchik A.A., “Postroenie parallelnykh form zadannoi shiriny iteratsionnykh algoritmov dlya realizatsii na matrichnykh protsessorakh”, Dokl. NAN Belarusi, 39:5 (1995), 17–20 | MR

[5] Krylov V.I., Bobkov V.V., Monastyrnyi P.I., Vychislitelnye metody, v. 2, Nauka, M., 1977 | MR

[6] Adutskevich E.V., Sobolevskii P.I., “Proektirovanie matrichnykh protsessorov s ispolzovaniem rasschepleniya i razdelnogo taimirovaniya makrooperatsii algoritma”, Trudy Instituta matematiki NAN Belarusi, 11 (2002), 22–33