Significant Publications

[1] J. Ahrens, B. Geveci, and C. Law, “Paraview: An End-user Tool for Large Data Visualization,” in The Visualization Handbook, pp. 717–731, Academic Press, 2005.

This paper describes the design and features of the ParaView visualization tool. The tool provides a graphical user interface for the creation and dynamic execution of visualization tasks. ParaView transparently supports the visualization and rendering of large data sets by executing these programs in parallel on shared or distributed memory machines. ParaView supports hardware-accelerated parallel rendering and achieves interactive rendering performance via level-of-detail techniques. The design balances and integrates a number of diverse requirements including the ability to handle large data, ease of use and extensibility by developers. This paper is a book chapter.

[2] J. Ahrens, K. Brislawn, K. Martin, B. Geveci, C. Law, and M. Papka, “Large-scale Data Visualization Using Parallel Data Streaming,” Computer Graphics and Applications, IEEE, vol.21, pp. 34–41, Jul/Aug 2001.

This paper describes the architectural changes to VTK/ParaView that enabled scalable parallel visualization and analysis. It presents an architectural approach to handling these visualization problems based on parallel data streaming. This enables visualizations on a parallel cluster that would normally require more storage/memory than is available while at the same time achieving high code reuse. Results from a variety of hardware and visualization configurations are discussed with data sizes ranging to a petabyte. The resulting architecture includes specific additions to support MPI, memory limit based streaming of both implicit and explicit topologies, translation of streaming requests between topologies, and passing data and pipeline control between shared, distributed, and mixed memory configurations. 

[3] J. Kniss, P. McCormick, A. McPherson, J. Ahrens, J. Painter, A. Keahey, C. Hansen, “Interactive Texture-based Volume Rendering for Large Data Sets,” Computer Graphics and Applications, IEEE, vol.21, pp. 52–61, Jul/Aug 2001. 

This paper describes a direct volume rendering technique that uses parallel graphics hardware, software-based compositing, and high-performance I/O to provide near-interactive display rates for time-varying, terabyte-sized data sets. It presents a foundational approach to using parallel graphics hardware to volume render massive datasets. A scalable, pipelined approach for rendering datasets too large for a single graphics card is presented. Our system, TRex, provides near-interactive display rates for time-varying, terabyte-sized uniformly sampled data sets and a low-latency platform for volume visualization in immersive environments. This paper has been cited 107 times. 

[4] J. Ahrens, K. Heitmann, M Petersen, J. Woodring, S. Williams, P. Fasel, C. Ahrens, C. Hsu, B. Geveci, "Verification of the Scientific Simulations via Hypothesis-Driven Comparative and Quantitative Visualization". IEEE Computer Graphics and Applications, Volume 30, Number 6, November/December 2010. 

This paper is highlighted because it embodies the successes of ParaView: innovative research, production delivery and solutions to real-world scientific problems. The research work, a visualization-assisted process for the verification of scientific simulation codes, is implemented in ParaView. We compare different cosmological and oceanographic simulations to reliably predict differences in simulation results. Our verification consists of the integration of an iterative hypothesis-verification process with comparative, feature, and quantitative visualization. We validate this process by verifying the results of real-world cosmology and oceanographic simulation verification problems. 

[5] C. Mitchell, J. Ahrens, J. Wang, “VisIO: Enabling Interactive Visualization of Ultra-Scale, Time Series Data via High-Bandwidth Distributed I/O Systems,” IPDPS 2011, pp. 68-79.

This paper represents recent work on next steps to integrate visualization with data intensive approaches. The I/O performance of parallel visualization applications can benefit from direct processor to local storage parallel reads such as those offered by using the Hadoop Distributed File System (HDFS). This paper describes modification to ParaView, called VisIO, to read, process and visualize simulation data stored as randomly distributed blocks on HDFS. We introduce a novel scheduling algorithm that helps to co-locate visualization processes on nodes with their requested data. Testing using VisIO integrated into ParaView was conducted using HDFS on TACC's Longhorn cluster. A representative plasma simulation dataset stored across 128 local storage/compute node pairs showed a 64.4% read performance improvement compared to the provided Lustre installation.