Shuhe Accelerates AI Model Service Deployment with Knative

Explore how Shuhe, a financial technology company, leverages Knative to accelerate AI model service deployment in this 45-minute conference talk. Learn about the challenges of frequent AI model iterations and multi-version deployments in financial business scenarios, and discover how Knative, an open-source serverless application architecture based on Kubernetes, addresses these issues. Gain insights into Shuhe's implementation, which has resulted in deploying over 500 AI model services, reducing resource costs by 60%, and shortening deployment cycles from 1 day to 0.5 days. Delve into practical aspects of deploying AI workloads with Knative, including expanding Serving elasticity capabilities, implementing Stable Diffusion, and adopting best practices for AI model services. This presentation offers valuable knowledge for organizations seeking to optimize AI service operations, reduce costs, and improve deployment efficiency in complex financial environments.

Shuhe Accelerates AI Model Service Deployment with Knative - Peng Li, Alibaba Cloud & Wenzhe Wei