ABSTRACT
We present a visualization system to assist designers of scheduling-based multi-threaded out-of-core algorithms. Our system facilitates the understanding and improving of the algorithm through a stack of visual widgets that effectively correlate the out-of-core system state with scheduling decisions. The stack presents an increasing refinement in the scope of both time and abstraction level; at the top of the stack, the evolution of a derived efficiency measure is shown for the scope of the entire out-of-core system execution and at the bottom the details of a single scheduling decision are displayed. The stack provides much more than a temporal zoom-effect as each widget presents a different view of the scheduling decision data, presenting distinct aspects of the out-of-core system state as well as correlating them with the neighboring widgets in the stack. This approach allows designers to to better understand and more effectively react to problems in scheduling or algorithm design.
As a case study we consider a global illumination renderer and show how visualization of the scheduling behavior has led to key improvements of the renderer's performance.
- Cox, P., Gauvin, S., and Rau-Chaplin, A. 2005. Adding parallelism to visual data flow programs. In Proceedings of the ACM Symposium on Software Visualization, ACM, 135--144. Google ScholarDigital Library
- Deelen, P., Van Ham, F., Huizing, C., and van de Wetering, H. 2007. Visualization of dynamic program aspects. International Workshop on Visualizing Software for Understanding and Analysis 0, 39--46.Google ScholarCross Ref
- Dutre, P., Bala, K., and Bekaert, P. 2002. Advanced Global Illumination. A. K. Peters, Ltd. Google ScholarDigital Library
- Hummel, S. F., Kimelman, D., Schonberg, E., Tennenhouse, M., and Zernik, D. 1997. Using program visualization for tuning parallel-loop scheduling. IEEE Concurrency 05, 1, 26--40.Google ScholarCross Ref
- Kienle, H. M., and Muller, H. A. 2007. Requirements of software visualization tools: A literature survey. International Workshop on Visualizing Software for Understanding and Analysis 0, 2--9.Google Scholar
- Moreta, S., and Telea, A. 2007. Visualizing dynamic memory allocations. International Workshop on Visualizing Software for Understanding and Analysis, 31--38.Google Scholar
- Mu, T., Tao, J., Schulz, M., and McKee, S. A. 2003. Interactive locality optimization on numa architectures. In Proceedings of the ACM Symposium on Software Visualization, ACM, 133. Google ScholarDigital Library
- Pharr, M., and Humphreys, G. 2004. Physically Based Rendering: From Theory to Implementation. Morgan Kaufmann Publishers Inc. Google ScholarDigital Library
- Reilly, M. 1990. Presentation tools for performance visualization: the m31 instrumentation experience. Proceedings of the Twenty-Third Annual Hawaii International Conference on System Sciences 1, 307--313.Google ScholarCross Ref
- Sawant, A. 2007. Diffarchviz: A tool to visualize correspondence between multiple representations of a software architecture. International Workshop on Visualizing Software for Understanding and Analysis 0, 121--128.Google ScholarCross Ref
- Shirley, P., and Morley, R. K. 2003. Realistic Ray Tracing. A. K. Peters, Ltd. Google ScholarDigital Library
- Smith, M. P., and Munro, M. 2002. Runtime visualisation of object oriented software. International Workshop on Visualizing Software for Understanding and Analysis 0, 81. Google ScholarDigital Library
- Summers, K. L., Caudell, T. P., Berkbigler, K., Bush, B., Davis, K., and Smith, S. 2004. Graph visualization for the analysis of the structure and dynamics of extreme-scale supercomputers. Information Visualization 3, 3, 209--222. Google ScholarDigital Library
- Wald, I., and Havran, V. 2006. On building fast kd-trees for ray tracing, and on doing that in O(N log N). In Proceedings of IEEE Symposium on Interactive Ray Tracing, 61--69.Google Scholar
- Zernik, D., Snir, M., and Malki, D. 1992. Using visualization tools to understand concurrency. IEEE Software 9, 3, 87--92. Google ScholarDigital Library
Index Terms
- Stacked-widget visualization of scheduling-based algorithms
Recommendations
Meta-model of information visualization based on Treemap
The interpretation and understanding of large quantities of data is a challenge for current information visualization methods. The visualization of information is important as it makes the appropriate acquisition of the information through the ...
Quantitative data visualization with interactive KDE surfaces
SCCG '10: Proceedings of the 26th Spring Conference on Computer GraphicsKernel density estimation (KDE) is an established statistical concept for assessing the distributional characteristics of data that also has proven its usefulness for data visualization. In this work, we present several enhancements to such a KDE-based ...
Dynamic switching of data visualization method for increased plotting scalability
NordiCHI '14: Proceedings of the 8th Nordic Conference on Human-Computer Interaction: Fun, Fast, FoundationalIn information visualization every method performs best in its own range of plotting density. When the available area for the visualization is unknown or when the data is supplied in packages of various sizes, the best suited method cannot be chosen in ...
Comments