{"id":6814,"date":"2017-12-12T23:34:55","date_gmt":"2017-12-12T22:34:55","guid":{"rendered":"http:\/\/www.mcomputers.cz\/?p=6814"},"modified":"2019-05-14T15:34:28","modified_gmt":"2019-05-14T14:34:28","slug":"nvidia-authorization-for-dgx-1","status":"publish","type":"post","link":"https:\/\/mcomputers.cz\/en\/2017\/12\/12\/nvidia-authorization-for-dgx-1\/","title":{"rendered":"NVIDIA authorization for DGX-1"},"content":{"rendered":"

Despite the very hectic end of the year, we were able to meet all the authorization conditions of NVIDIA for the area of machine learning<\/strong> and those for its flagship DGX-1, which boasts the designation of “the fastest way to machine learning”.\u00a0 We are thus building on our partnership with NVIDIA in the area Accelerated computing (GPGPU accelerators for the area of HPC in particular)<\/a>.<\/p>\n

\n
\"NVIDIA<\/a>

NVIDIA DGX<\/p><\/div>\n

Machine learning is in times of rapid development, and NVIDIA<\/a> is one of the world’s leaders in the area of hardware, and now, also software for accelerated machine learning.\u00a0 The dedicated NVIDIA DGX-1 server and the powerful workstation NVIDIA DGX Station are the result of its work. Both machines are based on NVIDIA’s most powerful iron and a fine-tuned software environment that includes a pre-installed and fine-tuned machine learning environment.\u00a0\u00a0 And it is these two technological jewels that can be purchased in the Czech and Slovak Republics thanks to our company.<\/p>\n

\"NVIDIA<\/a><\/p>\n

Acceleration of the learning phase against the traditional CPU computing environment and GPU environment vs. NVIDIA DGX-1.<\/p>\n

Hardware DGX-1<\/h3>\n

Let’s take a more detailed look at the NVIDA machines, first in terms of\u00a0 hardware.<\/p>\n

Parametr<\/th>DGX-1<\/th>DGX Station<\/th><\/tr>
GPUs<\/th>8\u00d7 Tesla V100<\/td>4\u00d7 Tesla V100<\/td><\/tr>
V\u00fdkon (GPU FP16)<\/th>1 PetaFLOPS<\/td>0,5 PetaFLOPS<\/td><\/tr>
GPU pam\u011b\u0165<\/th>128 GB celkem<\/td>64 GB celkem<\/td><\/tr>
CPU<\/th>2\u00d7 E5-2698 v4 2.2GHz (20 jader)<\/td>E5-2698 v4 2.2GHz (20 jader)<\/td><\/tr>
NVIDIA CUDA cores<\/th>40 960<\/td>20 480<\/td><\/tr>
NVIDIA Tensor cores<\/th>5 120<\/td>2 560<\/td><\/tr>
RAM<\/th>512 GB<\/td>256 GB<\/td><\/tr>
HDD<\/th>4\u00d7 1,92TB SSD<\/td>4\u00d7 1,92TB SSD<\/td><\/tr>
Network<\/th>Dual 10GbE, 4\u00d7 100Gb EDR Infiniband<\/td>Dual 10GbE<\/td><\/tr>
Maxim\u00e1ln\u00ed p\u0159\u00edkon<\/th>3 200 W<\/td>1 500 W<\/td><\/tr><\/tbody><\/table><\/div>