Adapting Kubernetes for Machine Learning Workflows
Explore how Bloomberg adapted Kubernetes to support diverse machine learning workloads in this keynote presentation from KubeCon + CloudNativeCon North America 2022. Discover the challenges and solutions in creating an internal machine learning platform that applies advanced AI and GPU-accelerated compute to various domains such as NLP, computer vision, time-series analysis, and personalization. Learn about the ecosystem of data connectors enabling rich collaboration, approaches to multi-tenancy, data security, and privacy, and methods for supporting heterogeneous compute requirements and hardware acceleration. Gain insights from senior engineers Ania Musial and Keith Laban as they share their experiences in leveraging Kubernetes as the foundation for Bloomberg's machine learning infrastructure.