An Introduction to Apache Pinot

on YouTube ↗54m

" summary

Apache Pinot is a real-time analytics DB that cuts per-query cost by orders of magnitude and targets user-facing OLAP needs rather than batch warehouses. It ingests directly from Kafka (its own consumer mapping partitions to segments), stores data in replicated segments on servers with brokers for scatter/gather, and achieves sub-100ms p99 at thousands of QPS in production (LinkedIn “who viewed my profile”, Uber Eats restaurant manager). Indexes include star-tree index, JSON and Lucene text, and the system scales toward petabyte scale.

AI-written summary · about

" chapters (12)

1. 0:00Introduction to Apache Pinot
2. 1:06Why Pinot matters now
3. 8:16History of analytics
4. 18:08Legacy tools and gaps
5. 23:13Pinot use cases
6. 28:49Pinot in modern stack
7. 35:32How Pinot fits systems
8. 40:14Pinot architecture
9. 42:02Data model and SQL
10. 43:26Pinot index types
11. 48:45Future of real-time analytics
12. 51:52Q&A

" related talks

• 49mIntroduction to Real-Time Analytics with Apache PinotTim Berglund · GOTO Chicago 2023
• 44mUSER-FACING, REALTIME ANALYTICS USING APACHE PINOTNeha Pawar · Crunch Conference 2021
• 37mCan Apache Pinot replace your OLTP database?Chinmay Soman · Berlin Buzzwords 2024
• 34mNext generation OLAP stack using Apache PinotChinmay Soman · Berlin Buzzwords 2022
• 47mReal-time analytics for your users with Apache PinotF. Teychene · DevOps D-Day 2022
• 33mPinot, Why Are You So Fast?Tim Berglund · Scale By the Bay 2023
• 46mOptimizing Speed and Scale of User-Facing Analytics w Kafka and PinotK Wolok, T Berglund · Voxxed Days Zürich 2022
• 47mReal-time Analytics for Your Users with Apache PinotFrançois Teychene · Devoxx France 2022