Название: cHARISMa

Тип системы: кластер

Веб-сайт: https://hpc.hse.ru/hardware/hpc-cluster

Место установки: Национальный исследовательский университет Высшая школа экономики

Год установки: 2019

Разработчики: Dell, Avilex, Hewlett Packard Enterprise, Институт системного программирования РАН (ИСП РАН)

Количество узлов: 54

Количество CPU: 108

Количество ядер CPU: 2936

Количество GPU-ускорителей: 166

Количество мультипроцессорных блоков GPU: 14584

Количество GPU-ядер/CUDA-ядер: 933376

Всего ядер: 17520

Архитектура:
узлов: 16 [2x Intel Xeon Gold 6152, 768 GB RAM; Acc: 4x NVIDIA Tesla V100]
узлов: 10 [2x Intel Xeon Gold 6152, 1536 GB RAM; Acc: 4x NVIDIA Tesla V100]
узлов: 6 [2x Intel Xeon Gold 6152, 768 GB RAM]
узлов: 2 [2x Intel Xeon Gold 6152, 768 GB RAM; Acc: 1x NVIDIA Tesla P40]
узлов: 3 [2x Intel Xeon Gold 6240R, 768 GB RAM; Acc: 4x NVIDIA Tesla V100]
узлов: 11 [2x Intel Xeon Gold 6248R, 384 GB RAM]
узлов: 6 [2x AMD EPYC 7702 , 1024 GB RAM; Acc: 8x NVIDIA A100]

Rpeak (Пфлоп/с): 2.027

Область применения: Наука и образование

Коммуникационная сеть: EDR Infiniband

Транспортная сеть: EDR Infiniband

Сервисная сеть: Gigabit Ethernet

Топология: FatTree

Семейство коммуникационной сети: InfiniBand

Linpack:
Результат (Rmax): 927.4 Тфлоп/с
Linpack Nmax: 797760
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 : 797760 NB : 384 PMAP : Row-major process mapping P : 18 Q : 9 PFACT : Left NBMIN : 2 NDIV : 2 RFACT : Left BCAST : 1ringM DEPTH : 0 SWAP : Spread-roll (long) L1 : no-transposed form U : transposed form EQUIL : yes 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 1.110223e-16 - Computational tests pass if scaled residuals are less than 16.0 2021-09-19 21:32:22.079 2021-09-19 21:39:20.969 ================================================================================ T/V N NB P Q Time Gflops -------------------------------------------------------------------------------- WR01L2L2 797760 384 18 9 418.90 8.080e+05 -------------------------------------------------------------------------------- ||Ax-b||_oo/(eps*(||A||_oo*||x||_oo+||b||_oo)*N)= 0.0001958 ...... 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. ================================================================================