Heterogeneous Computing

Education
SINTEF is a research institute without teaching activities by itself. Instead, we supervise students and teach courses at partner institutions. Our researchers have held courses in topics related to heterogeneous computing at three universities and colleges across Norway. Supervision of master and PhD students not only allows us to quickly target emerging technologies but is also central in our efforts to make heterogeneous computing available for Norwegian industry.

The Heterogeneous Computing group has supervised students from multiple universities in Norway, including the University of Oslo, Narvik University College, and the Norwegian University of Science and Technology. Our ongoing aim is to continue to supervise students at different levels, and we have several master's projects available. We have had many master's students through the years, and many graduees have continued to work at SINTEF, either as researchers, or as doctoral students. As a master student associated to us you will be integrated in our research group, including informal discussions at the foosball table.

 

Available Master's Projects

GPU maskin The projects we present here are suggestions, and show some of the width of our previous and ongoing master projects. If you have specific desires, please contact us for an informal discussion. Common to all the projects described is that they will begin with an introductory study of existing literature, and state of the art. The projects will continue with development, implementation and testing of (new) methods. All the projects require candidates with good experience with programming and/or numerical methods. The projects are further possible to tailor to the specific knowledge of the candidate. The projects will, in general, end up with results worthy of publication.

Depending on the requirements of the projects, the students will get access to the same hardware resources as our researchers, including, a 5-GPU computer, Playstation3, and Cell BE Blade Center. As we work in close cooperation with both industry and universities our activities often involve both state of the art technology and demanding customer requests. This enables us to provide project of both academic and industrial interest.

 

Virtualizing CUDA for C
Nvidias CUDA API is the preferred choice for many GPGPU computations. However, not all computers are CUDA-ready, and the correct configuration of drivers and compilers is a major headache. It would be beneficial to provide virtualized CUDA environments where a small pool of GPUs could be used by several guest operating systems with only a small performance penalty.

This project will include the development of a Xen driver that handles communication through the hypervisor and a userspace library that communicates with the driver.
Contact: Johan Seland 

Explicit schemes for evolutionary PDEs.
Shallow Water with Dry States SimulationExplicit schemes for evolutionary PDEs maps well to GPU hardware and high speedups have been demonstrated for applications to shallow water flows, gas dynamics, and astrophysics. The project can take several directions: (i) explore central difference schemes on triangular grids, (ii) investigate the use hierarchical parallelisation using multiple GPUs, (iii) use of higher-order (WENO) reconstructions for shallow water and/or gas dynamics.
Contact: Knut-Andreas Lie

Flow in porous media
SINTEF has for several years developed advanced computational methodologies for computing flow in porous media with particular emphasis on applications in reservoir simulation and subsurface storage of CO2. The project can take several directions: (i) use CUDA or OpenCL on programmable GPUs to accelerate streamline simulation: the solution of 1-D transport problems along streamlines and gravity lines and/or pvt and flash calculations; (ii) use of GPUs for linear algebra, in particular, algebraic multigrid and/or multiscale methods, (iii) build support for GPU accelleration into the MATLAB Reservoir Simulation Toolbox.
Contact: Knut-Andreas Lie

3D Volume Visualization
3D Electrocardiac visualizationVolume visualization is increasingly becoming more and more important. The use of computationally demanding algorithms, such as ray-tracing and ray-casting, has recently become feasible for realtime visualization.

Projects:

  1. Explore ray-tracing/ray-casting algorithms for efficient visualization of out-of-core datasets.
  2. Explore the possibilities for interactive flow visualization.

Visualization of smooth surfaces
Algebraic surface visualizationApplications ranging from computer games and simulators to industrial design rely on efficient rendering of smooth surfaces e.g. subdivision and B-Splines. Visualizing spline surfaces correctly and in real time is a vital part of all CAD software. Functionality available in new graphics hardware can be exploited to efficiently visualize such surfaces, both in real time, and with good quality.

Algebraic surfaces are often difficult to understand from their mathematical formulation. Visualization of such surfaces is important to broaden our understanding of such surfaces and their singularities.

Projects:

  1. Explore different volume-visualization for high-quality efficient visualization of algebraic surfaces.
  2. Explore the use of new graphics hardware for view-dependent spline surface visualization.
  3. Explore the possibilities to improve the smoothness of triangulations during rendering.

Contact: Tor Dokken

 

Interactive Ray-Tracing of Volumetric Data using the NVIDIA OptiX engine

More information to come.

Current Master/Project Students

  1. Lars Jahr Røine, University of Oslo
  2. Svein Atle Arnesen, University of Oslo
  3. Gagandeep Singh og Runar Heggelien Refsnæs, NTNU

Past Stutents

Post Doctoral

  • Talal Rahman (2008),  University of Bergen.

Doctoral

  • André R. Brodtkorb (2010), University of Oslo. Scientific Computing on Heterogeneous Architectures.
  • Jon M. Hjelmervik (2009), In cotutelle between University of Oslo and Grenoble INP. Heterogenous Computing with Focus on Mechanical Engineering.  
  • Johan S. Seland (2008),  University of Oslo. Smooth surface visualization using graphics hardware.

Master

  1. Erik Bjønnes (2010), University of Oslo, Variably sized filter kernels for GPU Accelerated Photon Mapping
  2. Asbjørn Bydal (2009), University of Agder, GPU-accelerated simulation of flow through a porous medium.
  3. Eirik Ulvik (2008), Narvik University College, Subdivision on the GPU.
  4. Kent Roger Bjørshol Fagerjord (2008), Narvik University College, Collision detection on the GPU
  5. Tatyana Lochehina (2008), Narvik University College, Collision detection on the GPU
  6. Håkon Fjukstad (2007), Narvik University College, Visualization of water flow in terrain
  7. Kato Griff Klæboe (2007), Narvik University College, GPU-based visualization of explosions in real time
  8. Ove Hermann Bastiansen (2007), Narvik University College, "Zoom" of visualization of complex geometry
  9. Xu Yihui (2007), Narvik University College, GPU-based interactive ray tracing of algebraic surfaces
  10. Martin Lilleeng Sætra (2007), University of Oslo, Solving systems of hyperbolic PDEs using multiple GPUs
  11. Lars Moastuen (2007), University of Oslo, Real-time simulation of the incompressible Navier-Stokes equations on the GPU
  12. Trygve Fladby (2007), University of Oslo, Efficient linear algebra on heterogeneous processors
  13. Hanne Moen (2007), University of Oslo, Wavelet transformation and efficient implementation on the GPU
  14. Andre Rigland Brodtkorb (2007), University of Oslo, A MATLAB interface to the GPU
  15. Olav Haugehåtveit (2006), Norwegian University of Science and Technology, Calculation of underwater sound fields on the GPU
  16. Andreas Eidissen (2006), Narvik University College, Realistic real-time simulation of waves from vessels
  17. Frank-Vegar Mortensen (2006), Narvik University College, A real-time system for creating and displaying ship wakes
  18. Ove Byberg (2006), Narvik University College, 3D terrain visualization based on geometric images using the GPU
  19. Andrei Kourinski (2005), Narvik University College, Vertex and fragment programs for image processing
  20. Bjørn Anders Ulsund (2005), Narvik University College, Expo-rational B-splines evaluated on the GPU
  21. Fredrik Eidissen (2005), Narvik University College, Course setup for GPGPU programming
  22. Gunhild Helene Strøm (2005), Narvik University College, Experiments with filters and transformations using vertex and fragment programs with applications in image processing.
  23. Hans-Børge Stormo (2005), Narvik University College, Vertex and fragment programs for curvature visualization
  24. Juri Bobrovsky (2005), Narvik University College, Recognition of geometric objects using the GPU
  25. Pål-Arve Nilsen (2005), Narvik University College, 3D fluid flow simulation with complex boundaries on the GPU
  26. Richard Dahl (2005), Narvik University College, Realistic simulations using expo-rational B-splines and GPU programming

Published December 2, 2008