Question about latency and response time

Elastic Stack Elasticsearch
1

Hi, I have 2 question regarding latency and response time in Elasticsearch.

  1. Why on different monitoring clients there are different response time?

For example in Kibana average request time is 700ms:

kibana
kibana1484×540 121 KB

In Graphana average request time is about ~300-500ms:

graphana
graphana896×353 107 KB

In Dynatrace it's about 100-150ms, though I'm not sure if its measure it correctly:

dynatrace
dynatrace3378×622 253 KB

  1. Why response time suddenly changed?

We've average response time in a cluster about 400ms, but after the new mapping update (we added 3 new fields), response time decreased to 100-250 ms, and kept it that way throughout the day.

decreased
decreased1132×398 28.4 KB

The next day, the response time increased again

increased
increased1140×374 11.9 KB

What could cause a decrease in response time after updating the mapping? And why did the response time rise back?

Thanks!

2

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