Multi-GPU Sort-Last Volume Visualization

In this paper, we propose an experimental study of an inexpensive off-the-shelf sort-last volume visualization architecture based upon multiple GPUs and a single CPU. We show how to efficiently make use of this architecture to achieve high performance sort-last volume visualization of large datasets. We analyze the bottlenecks of this architecture.We compare this architecture to a classical sort-last visualization system using a cluster of commodity machines interconnected by a gigabit Ethernet network. Based on extensive experiments, we show that this solution competes very well with a mid-sized PC cluster, while it significantly improves performance compared to a single standard PC.

[1]  Henry Fuchs,et al.  A sorting classification of parallel rendering , 1994, IEEE Computer Graphics and Applications.

[2]  Wolfgang Straßer,et al.  Advanced techniques for high-quality multi-resolution volume rendering , 2004, Comput. Graph..

[3]  Kenneth Moreland,et al.  Scalable Rendering on PC Clusters , 2000, IEEE Computer Graphics and Applications.

[4]  Bernd Hamann,et al.  Multiresolution techniques for interactive texture-based volume visualization , 1999, Proceedings Visualization '99 (Cat. No.99CB37067).

[5]  Kenneth I. Joy,et al.  Efficient Error Calculation for Multiresolution Texture-based Volume Visualization , 2003 .

[6]  Jean-Michel Dischler,et al.  Dynamic load balancing for parallel volume rendering , 2006, EGPGV '06.

[7]  Greg Humphreys,et al.  Chromium: a stream-processing framework for interactive rendering on clusters , 2002, SIGGRAPH.

[8]  Dirk Reiners,et al.  Sorted pipeline image composition , 2006, EGPGV '06.

[9]  Thomas Ertl,et al.  Hierarchical Visualization and Compression of Large Volume Datasets Using GPU Clusters , 2004, EGPGV.

[10]  Alan Heirich,et al.  Dynamic load balancing for parallel volume rendering , 2006 .

[11]  Stefan Eilemann,et al.  OpenGL multipipe SDK: a toolkit for scalable parallel rendering , 2005, VIS 05. IEEE Visualization, 2005..

[12]  Tom Duff,et al.  Compositing digital images , 1984, SIGGRAPH.

[13]  Wolfgang Straßer,et al.  Interactive rendering of large volume data sets , 2002, IEEE Visualization, 2002. VIS 2002..

[14]  John D. Owens,et al.  Distributed texture memory in a multi-GPU environment , 2006, GH '06.

[15]  Kwan-Liu Ma,et al.  Parallel volume rendering using binary-swap compositing , 1994, IEEE Computer Graphics and Applications.

[16]  Sheelagh Carpendale,et al.  A GPU Cluster Without the Clutter: A Drop-in Scalable Programmable-Pipeline with Several GPUs and Only One PC , 2006 .

[17]  Kwan-Liu Ma,et al.  SLIC: scheduled linear image compositing for parallel volume rendering , 2003, IEEE Symposium on Parallel and Large-Data Visualization and Graphics, 2003. PVG 2003..

[18]  Martin Kraus,et al.  High-quality pre-integrated volume rendering using hardware-accelerated pixel shading , 2001, HWWS '01.

[19]  Thomas Ertl,et al.  Level-of-Detail Volume Rendering via 3D Textures , 2000, 2000 IEEE Symposium on Volume Visualization (VV 2000).

[20]  Thomas Ertl,et al.  for Graphics-Hardware-based Cluster Systems , 2006 .

[21]  William M. Hsu.  Segmented ray casting for data parallel volume rendering , 1993 .

[22]  Michael Wand,et al.  A hardware architecture for multi-resolution volume rendering , 2005, HWWS '05.