核心概念
다중 테넌트, 다중 가속화기 시스템에서 테넌트별 QoS 관리를 위한 강화 학습의 중요성
統計
"The work of M. Palesi, who has contributed to the development of Sec. 3, has been supported by the Spoke 1 ”FutureHPC & BigData” of the Italian Research Center on High-Performance Computing, Big Data and Quantum Computing (ICSC)."
"The work of V. Catania, who has contributed to Sec. 4, has been supported by PNRR MUR project PE0000013-FAIR."
引用
"Each user, or “tenant”, commonly has unique quality expectations."
"This paper will delve into managing tenant-specific QoS within an online, firm real-time scheduling framework for Deep Neural Networks (DNNs) in a multi-tenant, multi-accelerator setting."