Pains and Gains: Lessons From Orchestrating Volumes for Remote Storage in Cloud Native AI Scenarios

Explore a 26-minute conference talk from KubeCon + CloudNativeCon where Zhihao Xu and Yashi Su from Alibaba Cloud share valuable lessons learned from orchestrating volumes for remote storage in cloud-native AI scenarios. Learn about the common challenges developers face when migrating AI workloads to Kubernetes, including unexpected resource consumption, unstable remote storage access, and difficulties in adding new data sources without disrupting workflows. Discover how Fluid, a CNCF project focused on data access and workload orchestration for AI applications, addresses these pain points to help users build more effective AI platforms on Kubernetes. Gain practical insights based on real user feedback that can help improve your cloud-native AI infrastructure.