Data Replication Options for Azure SQL PaaS Services
summary
Overview of concrete replication options and limits: SQL Data Sync uses DML triggers and a hub-and-member model to provide bi-directional synchronization with sync frequencies from 1 second to 30 days (demo shows a 2-second sync syncing US and Asia inventories), requires primary keys and provides eventual consistency via a local sync agent and private endpoints. Change Data Capture reads the transaction log into change tables (default retention 3 days). Change Tracking records which rows changed with lower overhead. Active geo-replication supports up to four async secondaries; read scale-out is available only in Premium/Business Critical/HyperScale.
AI-written summary · about
chapters (13)
- 0:00Intro
- 1:40Why is Data Replication Relevant
- 3:55SQL Data Sync
- 7:04Use Cases
- 7:43SQL Data Sync Demo
- 14:18Change Data Capture Demo
- 16:39Enable CDC
- 17:33CDC Use Cases
- 17:59Change Tracking
- 19:44Change Tracking vs CDC
- 20:30Column Tracking
- 21:21Active Geographication
- 23:14Scaling Out
related talks
31mData Replication in Azure SQL and Beyond
54mDesigning Azure Data Disaster Recovery Strategies
49mAnalytics in SQL Server and Azure SQL Database
50mAzure SQL DB Data Portability : Mirroring, CDC, Export/Import and DataSync
47mPerfecting business continuity for Azure SQL DB
01h 01mAzure SQL Database-Business Continuity During Disaster
50mAzure SQL Managed Instance Roadmap
31mSynchronous Geo-Replication over Azure Tables