Performance degradation of queries for AWS managed Elastic

We are using Elastic 5.5.2 and ingestion jobs are scheduled to run every day and each job lasts for around one hour. Once indexed, data is not mutated and the only action is to query data for our customers. Our data also has some geospatial fields. Our queries, perform some geospatial filtering, used to take around fifteen minutes per query on average few weeks ago but now it takes almost the double. After deep investigation on the issue, we found that the stats API for the indexes of the data, indexed after Feb 16, 2018, has the different response as compared to the previously indexed data.
You can find both responses in the attachment. The file named response_old contains the response of the stats API when called on the previously indexed data while response_new for the recently indexed data.

Any help will be appreciated.

Mukrram Ur Rahman

Please don't post pictures of text, they are difficult to read and some people may not be even able to see them.

Have you contacted AWS support?
How are you monitoring the cluster?

Yes, I did but did not hear anything from them yet. I did not find any better way to explain the problem other than posting images, let me know if there is any.


It's text, so paste that and format it as code.

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.