HTAP vs OLTP/OLAP
summary
HTAP claims to let a single system serve OLTP and OLAP simultaneously; implementations either keep row and column replicas synchronized or separate engines behind a router. Actual tools mentioned: MySQL, ClickHouse®, TiDB, Hadoop, Snowflake and cloud storage (S3) with NVMe and fast networks. Proven use cases include real-time dashboards and fraud detection; a TiDB cloud-native prototype reports promising results but partitioning data differently between row and column stores remains a practical pain point, so many teams still use MySQL + ClickHouse proxy hybrids.
AI-written summary · about
chapters (11)
- 0:00HTAP panel introduction
- 2:33Defining HTAP
- 5:29OLTP and OLAP combined
- 8:20Database architecture evolution
- 10:07HTAP enabling technologies
- 15:11HTAP use cases
- 16:16Real-time dashboards without HTAP
- 19:02Trade-offs in systems
- 23:36HTAP data partitioning
- 26:02Database trends and cycles
- 28:52Summary of HTAP
related talks
01h 13mHTAP made live — from the engineering perspective
41mHTAP with MariaDB Platform
01h 01mHTAP Workloads: Challenges and Solutions
49mHTAP with Azure Cosmos DB: Hybrid Transaction & Analytical Processing
47m10 Reasons Why the TiDB Database Should Power Your Next Development Project
58mUnderstanding Big Data for Software Engineers
38mOn the path to data Nirvana: Supporting OLTP and OLAP on Hadoop
38mKeynote | Data Warehouse, Data Lakes, What's Next?