TY - GEN
T1 - Thicket
T2 - 32nd International Symposium on High-Performance Parallel and Distributed Computing, HPDC 2023
AU - Brink, Stephanie
AU - McKinsey, Michael
AU - Boehme, David
AU - Scully-Allison, Connor
AU - Lumsden, Ian
AU - Hawkins, Daryl
AU - Burgess, Treece
AU - Lama, Vanessa
AU - Lüttgau, Jakob
AU - Isaacs, Katherine E.
AU - Taufer, Michela
AU - Pearce, Olga
N1 - Publisher Copyright:
© 2023 Owner/Author.
PY - 2023/8/7
Y1 - 2023/8/7
N2 - Thicket is an open-source Python toolkit for Exploratory Data Analysis (EDA) of multi-run performance experiments. It enables an understanding of optimal performance configuration for large-scale application codes. Most performance tools focus on a single execution (e.g., single platform, single measurement tool, single scale). Thicket bridges the gap to convenient analysis in multi-dimensional, multi-scale, multi-architecture, and multi-tool performance datasets by providing an interface for interacting with the performance data. Thicket has a modular structure composed of three components. The first component is a data structure for multi-dimensional performance data, which is composed automatically on the portable basis of call trees, and accommodates any subset of dimensions present in the dataset. The second is the metadata, enabling distinction and sub-selection of dimensions in performance data. The third is a dimensionality reduction mechanism, enabling analysis such as computing aggregated statistics on a given data dimension. Extensible mechanisms are available for applying analyses (e.g., top-down on Intel CPUs), data science techniques (e.g., K-means clustering from scikit-learn), modeling performance (e.g., Extra-P), and interactive visualization. We demonstrate the power and flexibility of Thicket through two case studies, first with the open-source RAJA Performance Suite on CPU and GPU clusters and another with a large physics simulation run on both a traditional HPC cluster and an AWS Parallel Cluster instance.
AB - Thicket is an open-source Python toolkit for Exploratory Data Analysis (EDA) of multi-run performance experiments. It enables an understanding of optimal performance configuration for large-scale application codes. Most performance tools focus on a single execution (e.g., single platform, single measurement tool, single scale). Thicket bridges the gap to convenient analysis in multi-dimensional, multi-scale, multi-architecture, and multi-tool performance datasets by providing an interface for interacting with the performance data. Thicket has a modular structure composed of three components. The first component is a data structure for multi-dimensional performance data, which is composed automatically on the portable basis of call trees, and accommodates any subset of dimensions present in the dataset. The second is the metadata, enabling distinction and sub-selection of dimensions in performance data. The third is a dimensionality reduction mechanism, enabling analysis such as computing aggregated statistics on a given data dimension. Extensible mechanisms are available for applying analyses (e.g., top-down on Intel CPUs), data science techniques (e.g., K-means clustering from scikit-learn), modeling performance (e.g., Extra-P), and interactive visualization. We demonstrate the power and flexibility of Thicket through two case studies, first with the open-source RAJA Performance Suite on CPU and GPU clusters and another with a large physics simulation run on both a traditional HPC cluster and an AWS Parallel Cluster instance.
KW - exploratory data analysis
KW - hpc
KW - multi-dimensional
KW - parallel profile
KW - performance analysis
UR - http://www.scopus.com/inward/record.url?scp=85169561769&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85169561769&partnerID=8YFLogxK
U2 - 10.1145/3588195.3592989
DO - 10.1145/3588195.3592989
M3 - Conference contribution
AN - SCOPUS:85169561769
T3 - HPDC 2023 - Proceedings of the 32nd International Symposium on High-Performance Parallel and Distributed Computing
SP - 281
EP - 293
BT - HPDC 2023 - Proceedings of the 32nd International Symposium on High-Performance Parallel and Distributed Computing
PB - Association for Computing Machinery, Inc
Y2 - 16 June 2023 through 23 June 2023
ER -