Never ETL again, thanks to Synapse Link. Really?
summary
Synapse Link claims to push OLTP rows to Azure Synapse in near‑real time but actually requires a self-hosted integration runtime, writes into an ADLS Gen2 landing zone and insists on a dedicated SQL pool (so indirect cost). It skips disallowed types (no XML or text or UDT columns), forbids replication/CDC overlap and disallows ad hoc ALTER/TRUNCATE — schema changes need reinitialization. Large updates trigger full table reloads and momentary empty targets; live demo struggled but deployment scripts are on GitHub.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:05Synapse Link for SQL Server
- 3:54Restrictions and limitations
- 6:03Pre-demo setup
- 9:13Synapse Link connections
- 12:35Demo attempt and troubleshooting
- 17:29Data update behavior and costs
- 18:53Q&A
related talks
19mNever ETL again, thanks to Synapse Link. Really?
23mSynapse Link to SQL, bring your on-prem data alive
50mWhen and how to migrate to Azure Synapse?
46mThe Ultimate Jigsaw - Architecting Azure Synapse Analytics
49mAzure Synapse Analytics
54mAn introduction to Azure Synapse Analytics!
24mSynapse Analytics
53mAzure Synapse Analytics: Evolving Azure SQL Data Warehouse