Running PostgreSQL and pgvector on Kubernetes: High-Performance AI Databases with Dell PowerFlex

Enterprises want AI capabilities without adding new database platforms or operational complexity. In this video, we explore how PostgreSQL and the pgvector extension run on Kubernetes, backed by Dell PowerFlex, to support both OLTP workloads and high‑performance vector search in a single system. Based on Dell Technologies White Paper H04651, this walkthrough highlights real benchmark results showing sub‑millisecond latency, over one million IOPS, and near‑linear scaling for PostgreSQL and pgvector workloads on Kubernetes. You’ll see how pgvector enables vector embeddings, similarity search, and approximate nearest neighbor indexing using SQL, with a comparison of IVF‑Flat and HNSW indexing for AI workloads such as semantic search, recommendations, and RAG pipelines. Topics covered: * PostgreSQL on Kubernetes * pgvector for vector search * IVF‑Flat vs HNSW indexing * Running OLTP and AI workloads together * PowerFlex CSI for Kubernetes storage * High‑performance database benchmarking Download the full Dell Technologies white paper (H04651): https://infohub.delltechnologies.com/en-us/t/postgresql-and-vector-database-extension-on-dell-powerflex-with-kubernetes/ This video is ideal for database administrators, platform engineers, and architects looking to make PostgreSQL AI‑ready without introducing a separate vector database.