End-to-End Study of Parallel Volume Rendering on the IBM Blue Gene/P
Authors:
- Tom PeterkaTPTom Peterka
- Hongfeng YuHYHongfeng Yu
- Robert RossRRRobert Ross
- Kwan-Liu MaKMKwan-Liu Ma
- Rob LathamRLRob Latham
- 1 September 2009
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 01903918,p. 566-573
Abstract
In addition to their role as simulation engines, modern supercomputers can be harnessed for scientific visualization. Their extensive concurrency, parallel storage systems, and high-performance interconnects can mitigate the expanding size and complexity of scientific datasets and prepare for in situ visualization of these data. In ongoing research into testing parallel volume rendering on the IBM Blue Gene/P (BG/P), we measure performance of disk I/O, rendering, and compositing on large datasets, and evaluate bottlenecks with respect to system-specific I/O and communication patterns. To extend the scalability of the direct-send image compositing stage of the volume rendering algorithm, we limit the number of compositing cores when many small messages are exchanged. To improve the data-loading stage of the volume renderer, we study the I/O signatures of the algorithm in detail. The results of this research affirm that a distributed-memory computing architecture such as BG/P is a scalable platform for large visualization problems.Keywords
This publication has 22 references indexed in Scilit:
- Collective communication: theory, practice, and experienceConcurrency and Computation: Practice and Experience, 2007
- Parallel unstructured volume rendering in ParaView2007
- COTS Cluster-based Sort-last Rendering: Performance Evaluation and Pipelined Implementation2006
- Optimization of Collective Communication Operations in MPICHThe International Journal of High Performance Computing Applications, 2005
- An Improved Algorithm for (Non-commutative) Reduce-Scatter with an Application2005
- A Performance and Scalability Analysis of the BlueGene/L Architecture2004
- Segmented ray casting for data parallel volume rendering2002
- An Extended Two‐Phase Method for Accessing Sections of Out‐of‐Core ArraysScientific Programming, 1996
- Parallel volume rendering using binary-swap compositingIEEE Computer Graphics and Applications, 1994
- A sorting classification of parallel renderingIEEE Computer Graphics and Applications, 1994