Название: Политехник - РСК Торнадо

Тип системы: кластер

Веб-сайт: http://scc.spbstu.ru

Место установки: Суперкомпьютерный центр,
Санкт‑Петербургский политехнический университет

Год установки: 2017

Разработчики: Группа компаний РСК

Количество узлов: 764

Количество CPU: 1528

Количество ядер CPU: 21992

Количество GPU-ускорителей: 128

Количество мультипроцессорных блоков GPU: 1840

Количество GPU-ядер/CUDA-ядер: 353280

Всего ядер: 23832

Архитектура:
узлов: 623 [2x Intel Xeon E5-2697v3, 64 GB RAM]
узлов: 56 [2x Intel Xeon E5-2697v3, 64 GB RAM; Acc: 2x NVIDIA Tesla K40]
узлов: 36 [2x Intel Xeon E5-2697v3, 128 GB RAM]
узлов: 8 [2x Intel Xeon E5-2697v3, 128 GB RAM; Acc: 1x NVIDIA K1]
узлов: 8 [2x Intel Xeon E5-2697v3, 128 GB RAM; Acc: 1x NVIDIA K2]
узлов: 3 [2x Intel Xeon E5-2697v3, 256 GB RAM]
узлов: 30 [2x Intel Xeon Platinum 8268, 192 GB RAM]

Rpeak (Пфлоп/с): 1.1487

Область применения: Исследования

Коммуникационная сеть: FDR Infiniband

Транспортная сеть: Gigabit Ethernet

Сервисная сеть: Gigabit Ethernet

Топология: FatTree

Семейство коммуникационной сети: InfiniBand

Linpack:
Результат (Rmax): 801.632 Тфлоп/с
Linpack Nmax: 2359296
Выдача Linpack: ================================================================================ HPLinpack 2.1 -- High-Performance Linpack benchmark -- October 26, 2012 Written by A. Petitet and R. Clint Whaley, Innovative Computing Laboratory, UTK Modified by Piotr Luszczek, Innovative Computing Laboratory, UTK Modified by Julien Langou, University of Colorado Denver ================================================================================ An explanation of the input/output parameters follows: T/V : Wall time / encoded variant. N : The order of the coefficient matrix A. NB : The partitioning blocking factor. P : The number of process rows. Q : The number of process columns. Time : Time in seconds to solve the linear system. Gflops : Rate of execution for solving the linear system. The following parameter values will be used: N : 2359296 NB : 192 PMAP : Column-major process mapping P : 16 Q : 191 PFACT : Left NBMIN : 4 NDIV : 2 RFACT : Left BCAST : 1ring DEPTH : 0 SWAP : Binary-exchange L1 : no-transposed form U : no-transposed form EQUIL : no ALIGN : 8 double precision words -------------------------------------------------------------------------------- - The matrix A is randomly generated for each test. - The following scaled residual check will be computed: ||Ax-b||_oo / ( eps * ( || x ||_oo * || A ||_oo + || b ||_oo ) * N ) - The relative machine precision (eps) is taken to be 1.111801e-16 - Computational tests pass if scaled residuals are less than 16.0 ================================================================================ T/V N NB P Q Time Gflops -------------------------------------------------------------------------------- WC00L2L4 2359296 192 16 191 10921.48 8.01632E+05 HPL_pdgesv() start time Sun Mar 15 10:11:47 2020 HPL_pdgesv() end time Sun Mar 15 13:13:48 2020 -------------------------------------------------------------------------------- ||Ax-b||_oo/(eps*(||A||_oo*||x||_oo+||b||_oo)*N)= 0.0007321 ...... PASSED ================================================================================ Finished 1 tests with the following results: 1 tests completed and passed residual checks, 0 tests completed and failed residual checks, 0 tests skipped because of illegal input values. -------------------------------------------------------------------------------- End of Tests. ================================================================================