Skip to content
By using AWS re:Post, you agree to the AWS re:Post Terms of Use
AWS re:Postre:Post
About re:Post

DMS Source Data and Latency spike nightly

0

I'm using a SQL Server on-prem source for a DMS to replicate to a SQL Server RDS. Every night, at about midnight, I'm seeing a spike of ~1.6M CDC incoming changes on my DMS task, but nothing is making anywhere near that many data changes on-prem. It takes about 8hr to get caught up, with a max latency of about 5hr. The latency hits its max about an hour or two before it's caught back up. Otherwise, it seems caught up for the remainder of the day. Also comparing the count of the PK column of one of the main tables, I'm about 65.6k records short in my RDS.

I don't have CDC on at the source db level or any of the source tables, but I thought I didn't need it because all tables being replicated have PKs

Topics
AnalyticsMigration & ModernizationDatabase
Tags
Microsoft SQL ServerAWS Database Migration ServiceAmazon Relational Database Service
Language
English
asked 2 years ago747 views
1 Answer
0

There can be multiple possible causes for DMS task latency and this can be resource and setting specific as well. I would recommend to go through these two articles first to check if there is any resource bottleneck or settings that can be tuned:

Troubleshooting DMS source latency: https://repost.aws/knowledge-center/dms-high-source-latency

Troubleshooting DMS target latency: https://repost.aws/knowledge-center/dms-high-target-latency

If above don't address the issue well, the best option would be raising a support case with support.

answered 2 years ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Relevant content