A new beta metric and library-based adaptive system mitigates GIL bottlenecks in edge AI, reaching 96.5% of optimal performance without manual tuning across tested workloads.
Reaching for the Sky: Maximizing Deep Learning Inference Throughput on Edge Devices with AI Multi-Tenancy
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Mitigating GIL Bottlenecks in Edge AI Systems
A new beta metric and library-based adaptive system mitigates GIL bottlenecks in edge AI, reaching 96.5% of optimal performance without manual tuning across tested workloads.