Lightning Talk: Bringing Multi-Tenancy to GPU-Based Workloads - Dario Tranchitella, Clastix
Автор: CNCF [Cloud Native Computing Foundation]
Загружено: 2024-03-19
Просмотров: 331
Описание:
Don't miss out! Join us at our next Flagship Conference: KubeCon + CloudNativeCon North America in Salt Lake City from November 12 - 15, 2024. Connect with our current graduated, incubating, and sandbox projects as the community gathers to further the education and advancement of cloud native computing. Learn more at https://kubecon.io
Lightning Talk: Bringing Multi-Tenancy to GPU-Based Workloads - Dario Tranchitella, Clastix
With the rise of AI, and machine learning back in the day, GPU workloads started having more traction in the Kubernetes ecosystem. Although they share some similarities, GPU-based workloads require an exclusive lock on the underlying storage, unless using specific hardware features, or by implementing a time-slicing strategy: Multi-Tenancy seems impracticable, in theory. In this lighting talk, I'm going to share how a GPU-based cluster could be consumed by external users achieving "a sort of" multi-tenancy, by combining the as-a-service paradigm thanks to hosted CNCF projects.
Повторяем попытку...
Доступные форматы для скачивания:
Скачать видео
-
Информация по загрузке:
![[Январь] Основной доклад: Использование MCP для раскрытия потенциала многоагентных структур](https://imager.clipsaver.ru/QYcf1RiKs4U/max.jpg)
