FastTrackGPU: Static-Analysis–Guided Analytical Modeling for Softcore GPUs

View at publisher Summarise with VIRAT
Authors: Chihyo Ahn, Shinnung Jeong, Liam Paul Cooper, Ruobing Han, Huanzhi Pu, Nicholas Parnenzini, Jisheng Zhao, Blaise Tine, Hyesoon Kim
Cited by: 0
References: 6

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

softGPUs make FPGA platforms practical data-parallel accelerators by providing a GPU-like programming model on top of an open RTL-to-software stack. However, their high configurability creates a large configuration space, making it difficult to identify workload-optimal designs under FPGA resource constraints. We address this challenge by proposing the first analytical model for softGPUs and four decision policies that aggressively prune the design space. Implemented in our FastTrackGPU framework, our lightweight analytical model driven by static analysis enables orders-of-magnitude faster configuration selection than Bayesian optimization while achieving up to 1.55× higher kernel throughput than area-greedy baselines.

Fields of Study & Subjects

Embedded Systems Design Techniques Parallel Computing and Optimization Techniques VLSI and FPGA Design Techniques

Publication details

Journal / source
IEEE Computer Architecture Letters
Volume
25
Issue
1
Pages
162–165
Publication date
1 January 2026
Language
EN