Next generation OLAP stack using Apache Pinot
summary
Apache Pinot is a distributed columnar OLAP engine that claims real-time user-facing analytics at scale: ingesting millions of events/sec and serving over 200,000 QPS with millisecond p99 latency. It implements local upsert bitmaps for accurate real-time mutations without global coordination, uses partitioning plus broker/server pruning and per-column indexes (inverted, range, geo, star-tree) to speed queries, and offers tiered storage on S3 plus a JSON index that is orders-of-magnitude faster than UDF parsing.
AI-written summary · about
chapters (9)
- 0:00Introduction to OLAP stack
- 1:02Evolving OLAP requirements
- 8:19Apache Pinot overview
- 12:32Pinot query speed optimizations
- 19:31Handling upserts in Pinot
- 23:06Hybrid storage architecture
- 26:42JSON data handling
- 28:05Pinot as modern OLAP engine
- 29:33Q&A session
related talks
44mUSER-FACING, REALTIME ANALYTICS USING APACHE PINOT
54mAn Introduction to Apache Pinot
49mIntroduction to Real-Time Analytics with Apache Pinot
38mReal-Time Analytics: Going Beyond Stream Processing with Apache Pinot
37mCan Apache Pinot replace your OLTP database?
46mOptimizing Speed and Scale of User-Facing Analytics w Kafka and Pinot
39mFaster Analytics for Fast Data with Apache Pinot and Flink SQL
47mReal-time analytics for your users with Apache Pinot