In the modern world one cannot imagine competing in science, business or industry without using powerful supercomputers capable to efficiently handle large scale calculations possibly on exponentially growing amounts of data. The focus of this track is to teach students how to efficiently and easily utilize modern supercomputing and Big Data architectures such as multi-core CPUs and GPUs for various applications in science and industry. Specifically, besides theoretical knowledge about parallel computing and Big Data, students will become familiar with standard parallel computing libraries (such as OpenMP, MPI, OpenACC and CUDA), Big Data frameworks (Hadoop), machine learning frameworks (such as Tensorflow), as well as visualization software (ParaView, Visit).
Additionally, students will be given a possibility to construct their own mini-supercomputer on the hands-on course, and learn how to maintain and administer it. Students will be given a chance of using Skoltech's world-class HPC facilities to learn typical methods and rules of working on the large-scale collectively used supercomputers.A successful graduate of this track will be able to:
- Use existing HPC and Big Data frameworks to successfully answer modern world challenges.
- Develop and optimize massively parallel computer codes capable to efficiently use modern architectures, such as CPUs and GPUs.
- Administer and maintain HPC and Big Data infrastructure.