Caliper+Thicket: Optimize HPC Simulations with Cross-Platform Exploratory Data Analysis @LivermoreLab
Caliper+Thicket: Optimize HPC Simulations with Cross-Platform Exploratory Data Analysis  @LivermoreLab
Uploaded May 2026 | Updated September 2026, 2 weeks ago
Caliper+Thicket is LLNL’s ubiquitous performance analysis software suite for optimizing HPC applications. Caliper and Thicket are complementary libraries that let users manually add lightweight performance instrumentation directly into their applications while deciding the granularity of information they want to collect and analyze during each application run. The resulting performance profiles reveal data of both breadth and depth, informing comparisons across runs and at scale.

Caliper+Thicket capabilities are particularly useful for large multiphysics simulation codes running on the world’s largest supercomputers, such as the codes and HPC systems deployed at LLNL, other scientific institutions, and other major computing centers. These libraries are open source, free, and available to the public on GitHub (github.com/llnl/Caliper, github.com/LLNL/thicket).

LLNL-VIDEO-2019589
Caliper+Thicket: Optimize HPC Simulations with Cross-Platform Exploratory Data AnalysisLLNL’s New Treatment to Combat FentanylLLNL 2023 Highlights
Lawrence Livermore National Laboratory |

Caliper+Thicket: Optimize HPC Simulations with Cross-Platform Exploratory Data Analysis

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER