Pixel-planes 5: a heterogeneous multiprocessor graphics system using processor-enhanced memories
SIGGRAPH '89 Proceedings of the 16th annual conference on Computer graphics and interactive techniques
A Sorting Classification of Parallel Rendering
IEEE Computer Graphics and Applications
InfiniteReality: a real-time graphics system
Proceedings of the 24th annual conference on Computer graphics and interactive techniques
Load balancing for multi-projector rendering systems
HWWS '99 Proceedings of the ACM SIGGRAPH/EUROGRAPHICS workshop on Graphics hardware
Hybrid sort-first and sort-last parallel rendering with a cluster of PCs
HWWS '00 Proceedings of the ACM SIGGRAPH/EUROGRAPHICS workshop on Graphics hardware
Sort-last parallel rendering for viewing extremely large data sets on tile displays
PVG '01 Proceedings of the IEEE 2001 symposium on parallel and large-data visualization and graphics
Chromium: a stream-processing framework for interactive rendering on clusters
Proceedings of the 29th annual conference on Computer graphics and interactive techniques
EGVE '03 Proceedings of the workshop on Virtual environments 2003
PVG '03 Proceedings of the 2003 IEEE Symposium on Parallel and Large-Data Visualization and Graphics
Validity of the single processor approach to achieving large scale computing capabilities
AFIPS '67 (Spring) Proceedings of the April 18-20, 1967, spring joint computer conference
Hi-index | 0.00 |
Most cluster-based rendering systems require at least one graphics processing unit (GPU) per node. However, there are many cluster systems which contain thin rack-mounted nodes that do not support GPUs. In this paper, we explore the utilization of such a system by combining it with a small commodity off-the-shelf tile rendering cluster. We call this a GPU-limited cluster because only those nodes responsible for rendering a tile contain GPUs. We examine the performance of sort-first, sort-middle, and sort-last approaches on a GPU-limited cluster with a six tile display system. Issues of cluster scalability and efficient utilization of nodes are also discussed.