Planetary scale fully managed artificial intelligence infrastructure services

The disclosure herein describes managing artificial intelligence (AI) workloads in a cloud infrastructure platform. A set of distributed infrastructure resources is integrated into the cloud infrastructure platform via a local support interface. AI workloads are received from a plurality of tenants,...

Full description

Saved in:
Bibliographic Details
Main Authors SVATANU, MIHAELA, KATIYAR AMIT, SHUKLA, DHARMA, K, SHARMA PRATEEK, NEHM RALF VON, MODI VIVEK A, SINGHAL, SANDEEP
Format Patent
LanguageChinese
English
Published 07.11.2023
Subjects
CALCULATING
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:The disclosure herein describes managing artificial intelligence (AI) workloads in a cloud infrastructure platform. A set of distributed infrastructure resources is integrated into the cloud infrastructure platform via a local support interface. AI workloads are received from a plurality of tenants, where the AI workloads include training workloads and inference workloads, and a subset of resources of a set of distributed infrastructure resources is assigned to the received AI workloads. The received AI workloads are scheduled for execution on the assigned subset of resources and they are executed on the assigned subset of resources based on the scheduling of the AI workloads. The described cloud infrastructure platform provides efficient and secure AI workload execution for many different tenants and enables flexible use of a wide variety of third-party infrastructure resources and first-party infrastructure resources. 本文中的公开内容描述了在云基础设施平台中管理人工智能(AI)工作负载。分布式基础设施资源集合经由本地支持接口集成到云基础设施平台中。从多个租户接收AI工作负载，其中AI工作负载包括
Bibliography:Application Number: CN202280022711