Efficient single- and multi-DNN inference using TensorRT framework

Vyacheslav Zhdanovskiy; Lev Teplyakov; Philipp Belyaev

doi:10.1117/12.3023487

3 April 2024 Efficient single- and multi-DNN inference using TensorRT framework

Vyacheslav Zhdanovskiy, Lev Teplyakov, Philipp Belyaev

Proceedings Volume 13072, Sixteenth International Conference on Machine Vision (ICMV 2023); 1307215 (2024) https://doi.org/10.1117/12.3023487
Event: Sixteenth International Conference on Machine Vision (ICMV 2023), 2023, Yerevan, Armenia

Abstract

In the recent years, there has been a significant growth of interest in real-world systems based on deep neural networks (DNNs). These systems typically incorporate multiple DNNs running simultaneously. In this paper we propose a novel approach of multi-DNN execution on a single GPU using multiple CUDA contexts and TensorRT, state-of-the-art DNN inference framework. We show that it can lead to more efficient scheduling of multiple DNNs, especially in case when a lightweight and a heavy DNNs are inferred together. We show that our approach can provide an almost 7x increase in the throughput of a lightweight DNN at the cost of neglible throughput drop of a heavy DNN, compared to the baseline. Moreover, we compare two ways of improving throughput of a single DNN by processing multiple images together: standard batching and implicit batching by processing multiple images simultaneously using several TensorRT execution contexts. We show that meanwhile standard batching outperforms implicit batching at larger batch sizes, implicit batching can provide up to 43% more throughput for a smaller DNN using smaller batch size.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Vyacheslav Zhdanovskiy, Lev Teplyakov, and Philipp Belyaev "Efficient single- and multi-DNN inference using TensorRT framework", Proc. SPIE 13072, Sixteenth International Conference on Machine Vision (ICMV 2023), 1307215 (3 April 2024); https://doi.org/10.1117/12.3023487

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available