Introducing StarRocks 4.1
What's new in StarRocks 4.1 — walked through feature by feature. → See PhoenixAI (managed StarRocks): https://bit.ly/4ycMopx → Questions about your workload? 👉 https://bit.ly/4zcAhJH StarRocks contributor and PhoenixAI PM Sida Shen walks through the StarRocks 4.1 release feature by feature — adaptive data distribution, large-capacity tablets, Apache Iceberg writes, cache observability, and new SQL — and the production problem each one solves, plus live Q&A. If you run or evaluate StarRocks or PhoenixAI — a real-time, open-source OLAP / MPP analytical database — this is a practical, engineer-to-engineer look at what 4.1 changes for your clusters. In this video: - Adaptive data distribution — automatic tablet split & merge, no more hand-tuning bucket counts (metadata-only). - Large-capacity tablets — less FE metadata, scheduling, and shuffle overhead. - Fast Schema Evolution v2 — metadata-only DDL on shared-data + materialized views. - Cache observability — cache metrics in the audit log and SQL profile. - Apache Iceberg & lakehouse — native DELETE/TRUNCATE, fewer small files, Iceberg V3 variant, incremental materialized views. - SQL — recursive CTEs, automatic skew-join handling, new & AI functions. - How PhoenixAI extends the 4.1 core: multi-warehouse, governance, encryption, cross-region DR, 24/7 support. Upgrade tip: start on 4.1.1, not 4.1.0. ------------------------------------------------------------------------------------------------------------------- Timestamps 0:00 Welcome & intro 0:52 StarRocks vs. PhoenixAI 1:57 Agenda & why 4.1 — the production pain points it targets 5:18 Storage: adaptive data distribution (range bucketing + auto tablet split/merge) 9:19 Large-capacity tablets 11:20 How Big Tablets Stay Fast 12:54 Fast Schema Evolution v2 (shared-data + materialized views) 13:48 Cache observability (audit log + SQL profile) 15:03 More shared-data storage updates (inverted index, tablet stats, snapshots) 16:51 Lakehouse & Iceberg: writes + table maintenance (delete, truncate, commit queue, small files) 20:51 Iceberg V3: variant data type 22:15 Incremental materialized views on Iceberg 24:00 Other Iceberg V3 & Lakehouse format updates 25:03 SQL & query: recursive CTEs 27:03 Skew join v2 & window-function upgrades 28:54 New functions, AI functions & upgrade notes (use 4.1.1) 30:11 PhoenixAI overview: enterprise layer on StarRocks + BYOC 31:16 StarRocks vs. PhoenixAI Cloud (isolation, governance, availability, caching, support) 33:21 New enterprise features with PhoenixAI (cross-region DR, cache stats, per-warehouse resource groups) 34:48 Live Q&A 35:15 Parquet load fails: "no value in non-nullable column" 36:11 Iceberg optimization (compacting manifests/data files) without Spark? 37:21 Do the 4.1 features apply to shared-nothing clusters? 37:58 How does AI_QUERY work with a table? (example) 38:28 Is commercial support available for StarRocks? 38:53 Any plans to support stored procedures? 39:17 Any plans to open-source multi-warehouse features? 39:40 Wrap-up & resources (4.1.1 release notes, PhoenixAI Cloud & Anywhere, 30-day trial) ------------------------------------------------------------------------------------------------------------------- Learn more at https://www.phoenixdata.ai/ Connect with us: LinkedIn: https://www.linkedin.com/company/phoenixai-data/ Twitter: https://x.com/phoenixdataai CelerData Website: https://www.phoenixdata.ai/ StarRocks GitHub: https://github.com/StarRocks/StarRocks StarRocks Website: https://www.starrocks.io/ Slack: https://starrocks.io/redirecting-to-slack #OpenSource #DataEngineering #DataInfrastructure #DataStack #DataAnalytics #RealTimeAnalytics #RealTimeData #OLAP #DataAnalyst #DataEngineer #DataInfrastructure #databaseprogramming #AgenticAnalytics #DataArchitecture #AIAnalytics #AgenticAnalytics #LLM #AIAgents #AnalyticsEngineering