Breakout Session

on YouTube ↗41m

" summary

PipelineDB is an open-source streaming SQL database built as a fork of PostgreSQL 9.4.4 that runs continuous SQL queries called continuous views which incrementally store only aggregate transition state, use a circular shared-memory stream buffer that blocks writers when full, and merge micro-batch partials with on-disk state via combiners. It ships probabilistic structures (top-k, cardinality with <1% error), offers a Kafka extension for durable offsets, claims typical throughput from 250 to over 1M events/sec, and is GPLv3 on GitHub.

AI-written summary · about

" chapters (11)

1. 0:00Introduction
2. 0:26PipelineDB
3. 3:34Why we built PipelineDB
4. 4:28Continuous SQL queries
5. 6:12Sliding window queries
6. 6:56Continuous queries
7. 7:59Adhoc queries
8. 11:23Streams
9. 18:26Combine
10. 26:14TopK
11. 32:09Questions

" related talks

• 48mUnlocking Real-time Insights: Enhancing Your Databases With Stream ProcessingYingjun Wu · Big Data Conference Europe 2023
• 33mStreaming DatabasesStan Zdonik · A Celebration of Mike Stonebraker (April 12, 2014)
• 04h 15mTopical Edition on Databases - Track 2Sandeep Joshi, Ranganadh Thata · Rootconf Topical Edition on Databases 2026
• 39mTimescaleDB: Re-engineering PostgreSQL as a time-series databaseErik Nordström · Øredev 2018
• 51mEmbrace, Extend, Evolve - Making PostgreSQL Work for a Cloud Native WorldKarthik Ranganathan · Distributed SQL Summit Asia 2023
• 10mBuilding High-Performance Financial Data PipelinesAnandan Dhanaraj · Conf42 Machine Learning 2025
• 49mPostgreSQL - the database-sister of PythonHarald Armin Massa · EuroPython 2011
• 30mThe database trends that are transforming your database infrastructure foreverPeter Zaitsev · LeadDev Berlin 2022