Spotlight
in
Workshop: Machine Learning for Systems
WarpDrive: An Agentic Workflow for Ninja GPU Transformations (Siva Hari, NVIDIA)
Sana Damani · Siva Kumar Sastry Hari · Mark Stephenson · Christos Kozyrakis
[
Abstract
]
[ Project Page ]
presentation:
Machine Learning for Systems
Sun 15 Dec 8:15 a.m. PST — 4:30 p.m. PST
[
OpenReview]
Sun 15 Dec 3:20 p.m. PST
— 3:30 p.m. PST
Sun 15 Dec 8:15 a.m. PST — 4:30 p.m. PST
Abstract:
Performance engineering for GPU-accelerated applications is challenging and time-consuming. We propose WarpDrive, a customizable LLM-driven performance analysis and optimization framework that automatically transforms and tests GPU applications. WarpDrive automates the optimization process using agents that analyze run time performance, create optimization plans, transform the code, and test for correctness. We demonstrate its effectiveness by customizing it to four different levels of optimization, including compiler options, compiler hints, function-level transformations, and application-level transformations.
Chat is not available.