Ultra-high resolution remote visualization with ParaView and SAGE

Research presented at UltraVis `09 during Supercomputing Conference 2009. paper link

Many of us don’t have physical access to high performance computing resources. This is why we do remote visualization where rendering is done by high performance render cluster located remotely and images are streamed to users over high-speed network.

The problem is you have to see the streamed visualizations on your tiny monitor. Today even a cheap digital camera can produce images which contains 11 Megapixels ( ~ 4000 x 2800 ). Your desktop monitor can’t display these images at their native resolution. What you’re going to see is either a small portion of the image at its native resolution or a scaled-down (sacrificing information in the image) version.  Imagine how big a visualized image of large scientific data. This research is about making the remote visualization (from ParaView) much more valuable by exploiting ultra-high resolution tiled-display wall technology.

ParaView is a scientific visualization tool that supports both remote visualization and parallel rendering. LambdaVision at EVL is the ultra-high resolution tiled-display wall, driven by graphics middleware called Scalable Adaptive Graphics Environment (SAGE) , providing 105 megapixel canvas. Combining ParaView’s parallel remote rendering with SAGE’s parallel streaming enables true ultra-high resolution remote visualization.

ParaView is one of the most widely used scientific tools that support parallel visualization of large scale data. The Scalable Adaptive Graphics Environment (SAGE) is a graphics middleware that enables real-time streaming of ultra-high resolution visual content from distributed visualization resources to scalable tiled displays connected by ultra-high-speed networks. Integrating these two technologies enables visualization of large-scale data at an extremely high resolution to be displayed on distantly located scalable tiled displays. The benefits, limitations, and future directions for this approach will be discussed.