Accelerate AI/ML workloads with Kubeflow and System Architecture

Presented by

Eduardo Aguilar Pelaez - Canonical & Srini Bala - Supermicro

About this talk

AI/ML model training is becoming more time consuming due to the increase in data needed to achieve higher accuracy levels. This is compounded by growing business expectations to frequently re-train and tune models as new data is available. The two combined is resulting in heavier compute demands for AI/ML applications. This trend is set to continue and is leading data center companies to prepare for greater compute and memory-intensive loads for AI. Getting the right hardware and configuration can overcome these challenges. In this webinar, you will learn: - Kubeflow and AI workload automation - System architecture optimized for AI/ML - Choices to balance system architecture, budget, IT staff time and staff training. - Software tools to support the chosen system architecture
Related topics:

More from this channel

Upcoming talks (4)
On-demand talks (410)
Subscribers (165545)
Get the most in depth information about the Ubuntu technology and services from Canonical. Learn why Ubuntu is the preferred Linux platform and how Canonical can help you make the most out of your Ubuntu environment.