Best practices for tuning Qlik Replicate CDC for big Bulk Insert / Update (Source: MSSQL, Target: MSSQL, CDC: MS-Replication)

[korsikov]

Hi team,
We use Qlik Replicate (Source: MS SQL, Target: MS SQL, CDC: MS-Replication).
Several times per day, the Source DB does massive bulk UPDATE + INSERT.
This causes Source latency to stay low, but Target latency grows to 20,000+ seconds.

Log shows huge batches (300k to 900k rows per bulk), finishing only due to bulk timeout.

 Question:

• What are recommended settings for Batch Apply Tuning and Change Processing Tuning in such a scenario?

• How to best handle massive bulk updates without latency overflow?

• Should we split batches or change commit frequency?

Any real-world best practices are appreciated!

[john_wang · Accepted solution]

Hello @korsikov ,

Thanks for reaching out to Qlik Community.

It's difficult to determine the root cause until we have a better understanding of where the bottleneck lies. There are several factors to consider, including the hardware and software configurations of the three servers involved (source SQL Server, Replicate Server, and target SQL Server), as well as the network throughput between them. I recommend reaching out to the Qlik Professional Services team for assistance in troubleshooting the performance issues.

Regards,

John.