Результаты исследований: Научные публикации в периодических изданиях › статья по материалам конференции › Рецензирование
Numerical hydrodynamics simulation of astrophysical flows at Intel Xeon Phi supercomputers. / Kulikov, I. M.; Chernykh, I. G.; Vorobyov, E. I. и др.
в: CEUR Workshop Proceedings, Том 1576, 2016, стр. 205-220.Результаты исследований: Научные публикации в периодических изданиях › статья по материалам конференции › Рецензирование
}
TY - JOUR
T1 - Numerical hydrodynamics simulation of astrophysical flows at Intel Xeon Phi supercomputers
AU - Kulikov, I. M.
AU - Chernykh, I. G.
AU - Vorobyov, E. I.
AU - Snytnikov, A. V.
AU - Weins, D. V.
AU - Moskovsky, A. A.
AU - Shmelev, A. B.
AU - Protasov, V. A.
AU - Serenko, A. A.
AU - Nenashev, V. E.
AU - Vshivkov, V. A.
AU - Rodionov, A. S.
AU - Glinsky, B. M.
AU - Tutukov, A. V.
PY - 2016
Y1 - 2016
N2 - In this paper we present the AstroPhi code for numerical simulation of astrophysical flows at Intel Xeon Phi supercomputers. The co-design of a computational astrophysics model is described. The parallel implementation and scalability tests of the AstroPhi code are presented. The results of simulation of interaction between intergalactic wind and a disk galaxy are provided. For AstroPhi code a 134x speed-up with one Intel Xeon Phi accelerator and 75% weak scaling efficiency on 224x Intel Xeon Phi accelerators was obtained. We obtained the peak performance on a 7168 × 1024 × 1024 mesh size using 53760 RSC PetaStream threads.
AB - In this paper we present the AstroPhi code for numerical simulation of astrophysical flows at Intel Xeon Phi supercomputers. The co-design of a computational astrophysics model is described. The parallel implementation and scalability tests of the AstroPhi code are presented. The results of simulation of interaction between intergalactic wind and a disk galaxy are provided. For AstroPhi code a 134x speed-up with one Intel Xeon Phi accelerator and 75% weak scaling efficiency on 224x Intel Xeon Phi accelerators was obtained. We obtained the peak performance on a 7168 × 1024 × 1024 mesh size using 53760 RSC PetaStream threads.
KW - High performance computing
KW - Intel Xeon Phi accelerators
KW - Numerical astrophysics
UR - http://www.scopus.com/inward/record.url?scp=84978468529&partnerID=8YFLogxK
M3 - Conference article
AN - SCOPUS:84978468529
VL - 1576
SP - 205
EP - 220
JO - CEUR Workshop Proceedings
JF - CEUR Workshop Proceedings
SN - 1613-0073
T2 - 10th Annual International Scientific Conference on Parallel Computing Technologies, PCT 2016
Y2 - 29 March 2016 through 31 March 2016
ER -
ID: 25330499