What type of study is this?

This is a Experimental Study study.

October 1, 2025ETRI JournalOpen Access

Exploring GPU sharing techniques for edge AI smart city applications

View Full Paper

Ask AI

Bookmark

Authors

SWShiao Y. WooUniversity of Louisville JYJihwan YeoKyungpook National University JKJin-Hong KimPohang Iron and Steel (South Korea)

Discussion

Member takes

Implication

This analysis shows the impact of gpu sharing techniques on throughput and resource utilization in edge AI applications, highlighting key performance tradeoffs.

Key Points

Optimizing gpu sharing can significantly enhance performance in edge ai applications, addressing scalability issues.
Threading minimizes memory usage and resource contention, while multiprocessing improves gpu utilization and reduces inference times.
Using the nvidia jetson agx orin platform, varying techniques were tested under simulated smart city conditions.
Understanding performance tradeoffs is crucial for developing efficient edge ai solutions in resource-constrained environments.

Cite This Study

Woo et al. (2025) studied this question.

synapsesocial.com/papers/68dd89defe798ba2fc497c40 https://doi.org/10.4218/etrij.2025-0065

View Full Paper

Ask AI

Bookmark

Implication

This analysis shows the impact of gpu sharing techniques on throughput and resource utilization in edge AI applications, highlighting key performance tradeoffs.

Exploring GPU sharing techniques for edge AI smart city applications

Discussion

Member takes

Implication

Key Points

Cite This Study

Exploring GPU sharing techniques for edge AI smart city applications

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider