Trainings on Application Development and Optimization

Trainings on Application Development and Optimization

Using of domain-specific applications sometimes requires a special knowledge or just a simple advice how to start, which could be useful even for professionals. Understanding of program runtime behavior – such as memory usage, logical structure, I/O influence etc. – is necessary for most optimization types and the effective utilization of supercomputing resources. Keeping both cases in mind, MSU arranges a number of trainings, intensive special courses and hand-on labs for users of MSU Supercomputing Center.

Specialists of Intel together with MSU-Intel Center carried out a series of trainings devoted to efficient programming with Intel HPC software tools. The key feature of the first event was meeting with Intel specialists who helped to understand and tune users own code during the training. Target programming languages included C and Fortran with MPI, OpenMP, or combination of both. The latest event was concentrated on Intel Xeon Phi features as well as Intel Trace Collector and Analyzer practice.

Accelrys Materials Studio is a popular modeling and simulation environment designed to allow researchers in materials science and chemistry to predict and understand the relationships of a materials atomic and molecular structure with its properties and behavior. After installing the Materials Studio on “Lomonosov” supercomputer, specialists from Accelrys were invited to conduct the training devoted to this toolkit. More than 60 MSU Supercomputing Center users took part in this 4-days event.

A number of trainings on performance analysis were arranged in MSU in the frame of joint Russian-European projects HOPSA (HOlistic Performance System Analysis) and APOS (Application Performance Optimization and Scalability). Accompanied with hand-on labs, trainings covered such toolkits as LWM, Score-P, Scalasca, Vampir, Extrae/Paraver, and Dimemas.

Also a special series of trainings by Rogue Wave specialists was arranged in MSU covered TotalView debugger and ThreadSpotter analyzer.

Advanced training for supercomputer system administrators is held in MSU by popular demand of other Russian supercomputing centers. Key topics of this training are supercomputers hardware and software stack, users and access control, batch systems, monitoring tools, compilers, and parallel libraries.

As a GPU Center of Excellence, MSU together with NVIDIA offers a number of trainings and courses on GPU programming technologies. Topics covered are GPGPU evolution, architecture of GPUs, CUDA, libraries (CUFFT, Thrust etc.), multi-GPU technologies, profiling, debugging and optimization, OpenACC.

User login