A fully hardware-managed scheduling architecture for AI accelerators

Abstract The proliferation of AI applications across diverse domains has driven the evolution of AI accelerators toward higher performance and energy efficiency. This paper addresses the critical challenge of task scheduling in AI accelerators by introducing a fully hardware-managed scheduling syste...

Descripció completa

Dades bibliogràfiques

Autors principals:	Libo Cheng, Liang Yang, Jian Shao, Xinyi Gu, Rong Qian, Xinwei Zhang, Donghao Li, Hongbin Wang, Xiaoqi Xia
Format:	Artigo
Idioma:	Inglês
Publicat:	Springer 2026-02-01
Col·lecció:	Journal of King Saud University: Computer and Information Sciences
Matèries:	AI accelerators Hardware-managed scheduling OCSR Computation-scheduling instructions Scheduling latency Throughput optimization
Accés en línia:	https://doi.org/10.1007/s44443-026-00513-z
Etiquetes:	Afegir etiqueta Sense etiquetes, Sigues el primer a etiquetar aquest registre!