Workshop on Programming and Performance Visualization Tools (ProTools 21)

Virtual (on HUBB)

November 14, 2021 (Sunday) 09:00 am - 12:00 pm Central Standard Time


               

Held in conjunction with SC21: The International Conference on High Performance Computing, Networking, Storage and Analysis, and
in cooperation with TCHPC: The IEEE Computer Society Technical Consortium on High Performance Computing

Background

Understanding program behavior is critical to overcome the expected architectural and programming complexities, such as limited power budgets, heterogeneity, hierarchical memories, shrinking I/O bandwidths, and performance variability, that arise on modern HPC platforms. To do so, HPC software developers need intuitive support tools for debugging, performance measurement, analysis, and tuning of large-scale HPC applications. Moreover, data collected from these tools such as hardware counters, communication traces, and network traffic can be far too large and too complex to be analyzed in a straightforward manner. We need new automatic analysis and visualization approaches to help application developers intuitively understand the multiple, interdependent effects that algorithmic choices have on application correctness or performance. The ProTools workshop combines two prior SC workshops: the Workshop on Visual Performance Analytics (VPA) and the Workshop on Extreme-Scale Programming Tools (ESPT).

The Workshop on Programming and Performance Visualization Tools (ProTools) intends to bring together HPC application developers, tool developers, and researchers from the visualization, performance, and program analysis fields for an exchange of new approaches to assist developers in analyzing, understanding, and optimizing programs for extreme-scale platforms.

Workshop Topics

  • Performance tools for scalable parallel platforms
  • Debugging and correctness tools for parallel programming paradigms
  • Scalable displays of performance data
  • Case studies demonstrating the use of performance visualization in practice
  • Program development tool chains (incl. IDEs) for parallel systems
  • Methodologies for performance engineering
  • Data models to enable scalable visualization
  • Graph representation of unstructured performance data
  • Tool technologies for extreme-scale challenges (e.g., scalability, resilience, power)
  • Tool support for accelerated architectures and large-scale multi-cores
  • Presentation of high-dimensional data
  • Visual correlations between multiple data source
  • Measurement and optimization tools for networks and I/O
  • Tool infrastructures and environments
  • Human-Computer Interfaces for exploring performance data
  • Multi-scale representations of performance data for visual exploration
  • Application developer experiences with programming and performance tools

Previous Workshops

The ProTools workshop combines two prior SC workshops: the Workshop on Visual Performance Analytics (VPA) and the Workshop on Extreme-Scale Programming Tools (ESPT).

Papers

Call for Papers

We solicit papers that focus on performance, debugging, and correctness tools for parallel programming paradigms as well as techniques and case studies at the intersection of performance analysis and visualization.

Papers must be submitted in PDF format (readable by Adobe Acrobat Reader 5.0 and higher) and formatted for 8.5” x 11” (U.S. Letter). Submissions should be a minimum of 6 pages and a maximum of 10 pages in the IEEE Conference format. The 10-page limit includes figures, tables, and references.

All papers must be submitted through the Supercomputing 2021 Linklings site. Submitted papers will be peer-reviewed and accepted papers will be published by IEEE TCHPC.

Dates

Important Dates

  • Submission deadline (extended): August 30 September 8, 2021 (AoE) (final deadline)
  • Notification of acceptance: September 24, 2021 (AoE)
  • Camera-ready deadline: October 8, 2021 (AoE)
  • Workshop: November 14, 2021
Program

Technical Program

The workshop will be held as a virtual event (on HUBB) on Sunday, November 14, 2021 from 09:00 am - 12:00 pm US Central Standard Time.

Time Session
9:00 - 9:05 Welcome and opening remarks. Abhinav Bhatele, Anthony Danalis, and David Boehme.
9:05 - 9:35 Differential Performance Analysis Workflow for Algorithmic Changes. Isabel Thärigen, Joachim Protze, Fabian Orland, and Marc-André Hermanns
9:35 - 10:05 Controlling the Runtime Overhead of Python Monitoring with Selective Instrumentation. Andreas Gocht-Zech, Alexander Grund, and Robert Schöne
10:05 - 10:30 Break
10:30 - 11:00 Measurement and Analysis of GPU-Accelerated OpenCL Computations on Intel GPUs. Aaron Cherian, Keren Zhou, Dejan Grubisic, Xiaozhu Meng, and John Mellor-Crummey
11:00 - 11:30 Profiling Large Taskflow Programs with Modern D3 and C++. Tsung-Wei Huang
11:30 - 12:00 Invited Talk: ROCm Performance Tools. Timour Paltashev (AMD)
Committees

Workshop Chairs

Abhinav Bhatele, University of Maryland, College Park, USA
David Boehme, Lawrence Livermore National Laboratory, USA
Anthony Danalis, University of Tennessee, Knoxville, USA

Program Committee

Andreas Knuepfer, TU Dresden
Brian Wylie, Juelich Supercomputing Centre
Gerhard Wellein, FAU Erlangen-Nuernberg
Jean-Baptiste Besnard, Paratools
John Mellor-Crummey, Rice University
Jonathan Madsen, Lawrence Berkeley National Lab
Judit Gimenez, Barcelona Supercomputing Centre
Karl Fuerlinger, LMU Munich
Kate Isaacs, University of Arizona
Kevin Huck, University of Oregon
Marc-Andre Hermanns, RWTH aachen
Markus Geimer, Juelich Supercomputing Centre
Martin Schulz, TU Munich
Michael Gerndt, TU Munich
Nathan Tallent, Pacific Northwest National Laboratory
Paul Rosen, University of South Florida
Stephanie Brink, Lawrence Livermore National Laboratory
Tanzima Islam, Texas State University