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.
I first began work as a primary researcher and engineer on Thicket during its inception in the Spring of 2022, and it has continued to be my main focus year-round even as I am writing this (Fall 2023). Getting the opportunity to work on Thicket has been critical to my career as I had the chance to start from scratch (writing the first lines in thicket.py) and be involved during one of the hardest parts of the project whilst trying to get it off the ground. I developed many core and auxiliary components of the project, getting the opportunity to problem solve and implement my own ideas. From there being able to see it through to its release and publishing of the paper, working with some of the first users, and it all coming together was amazing. In this process, I have become a better engineer, researcher, thinker, developer, and all around I am much more confident in my abilities than I was two years ago.
I have poured a lot of time into this project and am immensely proud of where it's at. That being said I did not nearly do it alone, as Thicket has had dozens of hands developing, advising, and testing it.