Introducing Instruction-Accurate Simulators for Performance Estimation of Autotuning Workloads

Accelerating Machine Learning (ML) workloads requires efficient methods due to their large optimization space. Autotuning has emerged as an effective approach for systematically evaluating variations of implementations. Traditionally, autotuning requires the workloads to be executed on the target ha...

Full description

Saved in:
Bibliographic Details
Published in:2025 62nd ACM/IEEE Design Automation Conference (DAC) pp. 1 - 7
Main Authors: Pelke, Rebecca, Bosbach, Nils, Reimann, Lennart M., Leupers, Rainer
Format: Conference Proceeding
Language:English
Published: IEEE 22.06.2025
Subjects:
Autotuning
cache optimization
Design automation
Estimation
gem5
Hardware
Machine learning
Optimization
Predictive models
Scalability
TVM
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Accelerating Machine Learning (ML) workloads requires efficient methods due to their large optimization space. Autotuning has emerged as an effective approach for systematically evaluating variations of implementations. Traditionally, autotuning requires the workloads to be executed on the target hardware (HW). We present an interface that allows executing autotuning workloads on simulators. This approach offers high scalability when the availability of the target HW is limited, as many simulations can be run in parallel on any accessible HW.Additionally, we evaluate the feasibility of using fast instruction-accurate simulators for autotuning. We train various predictors to forecast the performance of ML workload implementations on the target HW based on simulation statistics.Our results demonstrate that the tuned predictors are highly effective. The best workload implementation in terms of actual run time on the target HW is always within the top 3% of predictions for the tested x86, ARM, and RISC-V-based architectures. In the best case, this approach outperforms native execution on the target HW for embedded architectures when running as few as three samples on three simulators in parallel.
DOI:10.1109/DAC63849.2025.11133237