Visualization and Analysis of Parallel Dataflow Execution with Smart Traces

Most performance analysis tools focus on presenting an overload of details, with little application-dependent structure, and predefined statistical summaries. This makes the complex relations present in a parallel program not directly recognisable to the user, making the task of identifying performance issues more costly in both time and effort. In this work we investigate the requirements to create visualisations of execution traces of parallel programs modelled as dataflows. We propose the Smart Trace (ST) concept, to encode the structure of the data, and guide the construction of specialised visualizations. A visualization tool can then leverage the relationships in the data to automate a given analysis task. We show with examples the power and flexibility of visualisations we can create to address specific questions formulated about the analysis of the data, with emphasis in parallel dataflow traces.

[1]  Peter van der Stok,et al.  VIPER: a tool for the visualisation of parallel programs , 1995, Proceedings Euromicro Workshop on Parallel and Distributed Processing.

[2]  Gennady L. Andrienko,et al.  Exploratory analysis of spatial and temporal data - a systematic approach , 2005 .

[3]  Cláudio T. Silva,et al.  HyperFlow: A Heterogeneous Dataflow Architecture , 2012, EGPGV@Eurographics.

[4]  Bernd Mohr,et al.  Scalable performance visualization for data-parallel programs , 1994, Proceedings of IEEE Scalable High Performance Computing Conference.

[5]  N. Andrienko,et al.  Coordinated Multiple Views: a Critical View , 2007, Fifth International Conference on Coordinated and Multiple Views in Exploratory Visualization (CMV 2007).

[6]  Lucas Mello Schnorr,et al.  3D approach to the visualization of parallel applications and Grid monitoring information , 2008, 2008 9th IEEE/ACM International Conference on Grid Computing.

[7]  Martin Wattenberg,et al.  Stacked Graphs – Geometry & Aesthetics , 2008, IEEE Transactions on Visualization and Computer Graphics.

[8]  William Gropp,et al.  From Trace Generation to Visualization: A Performance Framework for Distributed Parallel Systems , 2000, ACM/IEEE SC 2000 Conference (SC'00).

[9]  Timothy Sherwood,et al.  Understanding and visualizing full systems with data flow tomography , 2008, ASPLOS.

[10]  Bryan Cantrill,et al.  Dynamic Instrumentation of Production Systems , 2004, USENIX Annual Technical Conference, General Track.

[11]  Eduard Gröller,et al.  World Lines , 2010, IEEE Transactions on Visualization and Computer Graphics.

[12]  Ching-Farn Eric Wu,et al.  Gantt Chart visualization for MPI and Apache multi-dimensional trace files , 2002, Ninth International Conference on Parallel and Distributed Systems, 2002. Proceedings..

[13]  Niklas Elmqvist,et al.  Graphical Perception of Multiple Time Series , 2010, IEEE Transactions on Visualization and Computer Graphics.

[14]  R. Daniel Bergeron,et al.  A model and a system for data-parallel program visualization , 1995, Proceedings Visualization '95.

[15]  Kwan-Liu Ma,et al.  Visualizing Large‐scale Parallel Communication Traces Using a Particle Animation Technique , 2013, Comput. Graph. Forum.

[16]  Michael T. Heath,et al.  Visualizing the performance of parallel programs , 1991, IEEE Software.

[17]  Guido Juckeland,et al.  High Resolution Program Flow Visualization of Hardware Accelerated Hybrid Multi-core Applications , 2010, 2010 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing.

[18]  Jürgen Döllner,et al.  Understanding complex multithreaded software systems by using trace visualization , 2010, SOFTVIS '10.

[19]  Matthias S. Müller,et al.  Developing Scalable Applications with Vampir, VampirServer and VampirTrace , 2007, PARCO.

[20]  Francisco F. Rivera,et al.  Software Tools for Performance Modeling of Parallel Programs , 2007, 2007 IEEE International Parallel and Distributed Processing Symposium.

[21]  Christoph W. Ueberhuber,et al.  Visualization of Scientific Parallel Programs , 1994, Lecture Notes in Computer Science.

[22]  Lucas Mello Schnorr,et al.  Towards Visualization Scalability through Time Intervals and Hierarchical Organization of Monitoring Data , 2009, 2009 9th IEEE/ACM International Symposium on Cluster Computing and the Grid.

[23]  Dieter Schmalstieg,et al.  Comparative Analysis of Multidimensional, Quantitative Data , 2010, IEEE Transactions on Visualization and Computer Graphics.

[24]  Allen D. Malony,et al.  The Tau Parallel Performance System , 2006, Int. J. High Perform. Comput. Appl..

[25]  Jarke J. van Wijk,et al.  Trace Visualization Using Hierarchical Edge Bundles and Massive Sequence Views , 2007, 2007 4th IEEE International Workshop on Visualizing Software for Understanding and Analysis.

[26]  Lucas Mello Schnorr,et al.  DIMVisual: Data Integration Model for Visualization of Parallel Programs Behavior , 2006, Sixth IEEE International Symposium on Cluster Computing and the Grid (CCGRID'06).

[27]  Tobias Höllerer,et al.  behaviorism: a framework for dynamic data visualization , 2010, IEEE Transactions on Visualization and Computer Graphics.

[28]  Danny Holten,et al.  Hierarchical Edge Bundles: Visualization of Adjacency Relations in Hierarchical Data , 2006, IEEE Transactions on Visualization and Computer Graphics.

[29]  Wolfgang E. Nagel,et al.  High performance event trace visualization , 2005, 13th Euromicro Conference on Parallel, Distributed and Network-Based Processing.

[30]  Lucy T. Nowell,et al.  ThemeRiver: visualizing theme changes over time , 2000, IEEE Symposium on Information Visualization 2000. INFOVIS 2000. Proceedings.

[31]  J.C. Roberts,et al.  State of the Art: Coordinated & Multiple Views in Exploratory Visualization , 2007, Fifth International Conference on Coordinated and Multiple Views in Exploratory Visualization (CMV 2007).

[32]  Ewing Lusk,et al.  Performance analysis of MPI programs , 1994 .

[33]  James Roberts.  TraceVis: An Execution Trace Visualization Tool , 2004 .

[34]  Eileen Kraemer,et al.  The Visualization of Parallel Systems: An Overview , 1993, J. Parallel Distributed Comput..