Lobachevsky State University of Nizhni Novgorod (UNN)
has been designated a 2011 NVIDIA CUDA Research Center (CRC) based on the
vision, quality, and impact of its research leveraging CUDA technology. NVIDIA
recognizes and fosters collaboration with research groups at universities and
research institutes that are expanding the frontier of massively parallel
computing using graphics processing units(GPUs).
UNN is one of the leading Russian higher educational
institutions with high professional reputation. In October 2009 UNN became one
of a 12 Russian Universities to be awarded the title of the National Research
University. One of the priority development trends of UNN as a national research
university is the high performance and supercomputing technologies in the
development of systems and application software for multiprocessor computational
systems.
The strategic purpose of setting up CUDA Research Center in
UNN is to contribute to the wide use of perspective high performance computer
systems on the basis of graphics processors for solving the current
scientific-technical issues for organizations, enterprises and companies in
education, science, industry and business in Privolzhsky Federal District and
Russia in whole. Our goal is to broaden GPU computing in bioimaging, optical
tomography, radiotherapy, molecular dynamics and bio- and nano-engineering,
modeling of the brain functions and the brain-computer interfaces, computations
of laser accelerators of charged particles for radiotherapy, protonography and
inertial thermonuclear fusion.
Participants
Professor Victor Gergel is a dean of the Faculty
of Computational Mathematics and Cybernetics at Lobachevsky State University of
Nizhni Novgorod (UNN). He received his D.Sc. degree in CAD systems design. His
research interests include mathematical models, methods and decision-making
systems (the problems of global and multiobjective optimization, pattern
recognition and classification), software engineering and development of complex
software applications, mathematical models, methods and software systems for
parallel computing on multiprocessor computer systems (clusters). He was a
co-director of Russian Foundation for Basic Research project on developing
methods and software for computer-aided decision making. Leader of project for
developing computer software for decision making (this software has been
presented at the World International Exhibition CeBIT 95 & 07, Hannover,
Germany). Co-manager of the project of the Program of Russian-Hungarian Research
and Technological Cooperation and of the joint Russian-Dutch project. Co-manager
the research program on Perspective Information Technology in Computer-Aided
Modeling and Research.
Vadim Turlapov
is a professor of the Faculty of Computational Mathematics and Cybernetics at
Lobachevsky State University of Nizhni Novgorod (UNN), D.Sc. of Technics. He has
been teaching a course in Computer Graphics and Computer Architecture at UNN
since 2002. In 2006, he headed a Laboratory of Computer Graphics. In 2010, he
organized the All-Russian Scientific School «Computer vision, 3D modeling and
computer graphics». He is one of the founders of the Russian Community of
Computer Graphics Specialists GRAPHICON, established in 2010. The scientific
interests of his laboratory are Global Illumination, Rendering of Natural
Phenomena, Scientific Visualization in the context of real-time computing on the
GPU.
Konstantin Barkalov is a senior lecturer of the
Faculty of the Computational Mathematics and Cybernetics at Lobachevsky State
University of Nizhni Novgorod (UNN). He received a Ph.D. in Physics and
Mathematics. He participated in the TEMPUS-Tacis project “Statistics for
Regional Needs: Structures and Programs”. Member of Dutch-Russian research
project “High-speed computing in global optimization: sequential and parallel
environment” (NWO 047.016.014). Co-manager of the Russian Foundation for Basic
Research project on developing of methods and software for computer-aided
decision making “Multiextremal models and efficient methods of optimal choice”
(RFBR № 07-01-00467).
Denis Bogolepov is a post-graduate student of the
Faculty of Computational Mathematics and Cybernetics at Lobachevsky State
University of Nizhni Novgorod (UNN). He received his M.Sc. in Computer Science
in 2009. His research interests include General-Purpose computation on GPUs
(GPGPU) and realistic image synthesis. At UNN he teaches disciplines related to
computer graphics and HPC (OpenCL-related part of “Models and methods of HPC”
course), and participates in educational projects such as Intel Schools. His
current project is dedicated to real-time global illumination of dynamic scenes.
To speed up the computation GPU is used for all phases of rendering (including
building of acceleration structure).
Nikolay Gavrilov is a post-graduate student of the
Faculty of Computational Mathematics and Cybernetics at Lobachevsky State
University of Nizhni Novgorod (UNN). He received his M.Sc. in Computer Science
in 2009. His research interests include volume rendering for visualization of
scientific and medical data. His current project is dedicated to real-time
visualization of CT and MRI tomograms. His program provides several
visualization techniques, such as Shaded Direct Volume Rendering via the 1D- or
2D- transfer functions, multiple semi-transparent discrete iso-surfaces
visualization, MIP, MIDA, etc. He uses GPU-based algorithms with some
modifications to improve quality and rendering performance.
Dmitry Sopin is a graduate student of the
Faculty of Computational Mathematics and Cybernetics at Lobachevsky State
University of Nizhni Novgorod (UNN). He received his B.Sc. in Computer Science
in 2010. His research interests include General-Purpose Computations on GPUs
(GPGPU) and realistic image synthesis. He is an expert in graphics design and
development of tools for GPU. His current project is dedicated to real-time
global illumination for dynamic scenes. The current implementation performs all
rendering phases (including building of acceleration structure) on GPU and is
based on cross-platform standards OpenGL/OpenCL, therefore supports various
software and hardware platforms.
Sergey Bastrakov is a graduate student of the Faculty
of Computational Mathematics and Cybernetics at Lobachevsky State University of
Nizhni Novgorod (UNN). He received his B.Sc. in applied mathematics and computer
science (2010). He has 3.5 years experience in field of HPC and GPGPU: research
in effective implementation of global optimization and financial mathematics
algorithms for modern CPU and GPU hardware. His current research is related to
developing high performance plasma simulation code for heterogeneous (CPU + GPU)
clusters in collaboration with the Institute of Applied Physics of the Russian
Academy of Sciences. At UNN he teaches CUDA-related part of “Models and methods
of HPC” course.