Performance analysis is an integral part of developing and optimizing parallel applications for high performance computing (HPC) platforms. Hierarchical data from different sources is typically available to identify performance issues or anomalies. Some hierarchical data such as the calling context can be very large in terms of breadth and depth of the hierarchy. Classic tree visualizations quickly reach their limits in analyzing such hierarchies with the abundance of information to display. In this position paper, we identify the challenges commonly faced by the HPC community in visualizing hierarchical performance data, with a focus on calling context trees. Furthermore, we motivate and lay out the bases of a visualization that addresses some of these challenges.
%0 Conference Paper
%1 bergel2019visual
%A Bergel, Alexandre
%A Bhatele, Abhinav
%A Boehme, David
%A Gralka, Patrick
%A Griffin, Kevin
%A Hermanns, Marc-André
%A Okanović, Dusan
%A Pearce, Olga
%A Vierjahn, Tom
%B Programming and Performance Visualization Tools
%C Cham
%D 2019
%E Bhatele, Abhinav
%E Boehme, David
%E Levine, Joshua A.
%E Malony, Allen D.
%E Schulz, Martin
%I Springer International Publishing
%K sfb716-d3 myown from:patrickgralka vis(us) visus:gralkapk
%P 233--249
%T Visual Analytics Challenges in Analyzing Calling Context Trees
%X Performance analysis is an integral part of developing and optimizing parallel applications for high performance computing (HPC) platforms. Hierarchical data from different sources is typically available to identify performance issues or anomalies. Some hierarchical data such as the calling context can be very large in terms of breadth and depth of the hierarchy. Classic tree visualizations quickly reach their limits in analyzing such hierarchies with the abundance of information to display. In this position paper, we identify the challenges commonly faced by the HPC community in visualizing hierarchical performance data, with a focus on calling context trees. Furthermore, we motivate and lay out the bases of a visualization that addresses some of these challenges.
%@ 978-3-030-17872-7
@inproceedings{bergel2019visual,
abstract = {Performance analysis is an integral part of developing and optimizing parallel applications for high performance computing (HPC) platforms. Hierarchical data from different sources is typically available to identify performance issues or anomalies. Some hierarchical data such as the calling context can be very large in terms of breadth and depth of the hierarchy. Classic tree visualizations quickly reach their limits in analyzing such hierarchies with the abundance of information to display. In this position paper, we identify the challenges commonly faced by the HPC community in visualizing hierarchical performance data, with a focus on calling context trees. Furthermore, we motivate and lay out the bases of a visualization that addresses some of these challenges.},
added-at = {2019-05-02T14:12:49.000+0200},
address = {Cham},
author = {Bergel, Alexandre and Bhatele, Abhinav and Boehme, David and Gralka, Patrick and Griffin, Kevin and Hermanns, Marc-Andr{\'e} and Okanovi{\'{c}}, Du{\v{s}}an and Pearce, Olga and Vierjahn, Tom},
biburl = {https://puma.ub.uni-stuttgart.de/bibtex/24fbcfd6f34527c7dee6a432737c3c825/visus},
booktitle = {Programming and Performance Visualization Tools},
editor = {Bhatele, Abhinav and Boehme, David and Levine, Joshua A. and Malony, Allen D. and Schulz, Martin},
interhash = {5045973d25b5bbb51abd979adcccfd7d},
intrahash = {4fbcfd6f34527c7dee6a432737c3c825},
isbn = {978-3-030-17872-7},
keywords = {sfb716-d3 myown from:patrickgralka vis(us) visus:gralkapk},
pages = {233--249},
publisher = {Springer International Publishing},
timestamp = {2019-05-02T12:12:49.000+0200},
title = {Visual Analytics Challenges in Analyzing Calling Context Trees},
year = 2019
}