Topological Analysis And Visualization Of Cyclical Behavior In Memory Reference Traces

We demonstrate the application of topological analysis techniques to the rather unexpected domain of software visualization. We collect a memory reference trace from a running program, recasting the linear flow of trace records as a high-dimensional point cloud in a metric space. We use topological persistence to automatically detect significant circular structures in the point cloud, which represent recurrent or cyclical runtime program behaviors. We visualize such recurrences using radial plots to display their time evolution, offering multi-scale visual insights, and detecting potential candidates for memory performance optimization. We then present several case studies to demonstrate some key insights obtained using our techniques.

ANM Imroz Choudhury, Bei Wang, Paul Rosen, and Valerio Pascucci. Topological Analysis And Visualization Of Cyclical Behavior In Memory Reference Traces. IEEE Pacific Visualization Symposium, 2012.

@inproceedings{choudhury2012topological,
  title = {Topological Analysis and Visualization of Cyclical Behavior in Memory
    Reference Traces},
  author = {Choudhury, ANM Imroz and Wang, Bei and Rosen, Paul and Pascucci, Valerio},
  booktitle = {IEEE Pacific Visualization Symposium},
  series = {PacificVis},
  pages = {9--16},
  year = {2012},
  keywords = {automatic circular structure detection; cyclical behavior visualization;
cyclical runtime program behaviors; high-dimensional point cloud;
memory performance optimization; memory reference traces; metric
space; multiscale visual insights; radial plots; recurrence visualization;
software visualization; topological analysis; trace record linear
flow recasting; data visualisation; performance evaluation; storage
management; topology;},
  abstract = {We demonstrate the application of topological analysis techniques to the
    rather unexpected domain of software visualization. We collect a memory reference trace
    from a running program, recasting the linear flow of trace records as a high-dimensional
    point cloud in a metric space. We use topological persistence to automatically detect
    significant circular structures in the point cloud, which represent recurrent or
    cyclical runtime program behaviors. We visualize such recurrences using radial plots to
    display their time evolution, offering multi-scale visual insights, and detecting
    potential candidates for memory performance optimization. We then present several case
    studies to demonstrate some key insights obtained using our techniques.}
}