MTS has launched MTS GROM supercomputer

The supercomputer plays a key role in training of artificial intelligence algorithms. MTS GROM's computing power makes it possible to build more complex models, test by times more hypotheses, and train neural networks dozens of times faster than before. This directly affects the quality of products developed by MTS Artificial Intelligence Center and the speed of their market launch.

In the future, high-performance MTS GROM capacities will become available to Russian companies in #CloudMTS provider's product package. This will enable large businesses to significantly reduce the time required to develop and implement artificial intelligence and big data-based projects. Startups will be able to reduce costs of high-performance computing for speech analysis, video processing and facial recognition by hundreds of times. For example, fintech companies can use MTS GROM resources to develop remote identification solutions or to assess credit risks.

The supercomputer is deployed by #CloudMTS cloud provider in its proprietary MTS data center based on integrated NVIDIA DGX SuperPOD solution. This solution ensures not only the high speed of implementing supercomputing systems, but scalable performance for dealing with the most complex AI tasks and high-performance computing as well.

MTS GROM is the most powerful supercomputer not only in the Russian, but in the European and American telecom as well. It ranks 241st in the world ranking of high-performance computers.

MTS forms a digital ecosystem for Russian companies. Based on MTS GROM, we will develop the most promising technologies related to artificial intelligence, big data analysis, and cloud solutions. MTS GROM supercomputer is aimed at improving a company's technological capabilities, as well as forming the basis of future services that businesses and scientific organizations will be able to receive in #CloudMTS with significant cost savings, said Oleg Motovilov, the Director of MTS cloud business.

About MTS GROM supercomputer

• The supercomputer makes it possible to run a mass of simultaneous calculations on all computing nodes (servers). The solution is implemented on the basis of NVIDIA DGX A100 hardware and software platform with NVIDIA A100 GPUs with total 320 GB of memory.
• Computing nodes are connected by the high-speed Infiniband network. It is a commutable computer network that is used in high-performance computing and features high bandwidth and low delay time.
• MTS GROM computing cluster is equipped with NVMe technology-based NetApp data storage system, which provides superfast interaction between computing nodes and the data storage system to reduce the AI models training time.
• In the future, you will be able to use MTS GROM capacities by leaving a request on #CloudMTS website.