# Affect of disparate CPU speeds on cluster

**URL:** <https://discuss.elastic.co/t/affect-of-disparate-cpu-speeds-on-cluster/222962>\
**Category:** Elasticsearch\
**Created:** [March 10, 2020, 3:10pm UTC](https://discuss.elastic.co/t/affect-of-disparate-cpu-speeds-on-cluster/222962 "2020-03-10T15:10:45Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![HockeyFan0000](https://avatars.discourse-cdn.com/v4/letter/h/b4bc9f/32.png) [@HockeyFan0000](https://discuss.elastic.co/u/HockeyFan0000)\
**Post date:** [March 10, 2020, 3:10pm UTC](https://discuss.elastic.co/t/affect-of-disparate-cpu-speeds-on-cluster/222962/1 "2020-03-10T15:10:45Z")

</div>

I've replaced some ESX hosts with new ones and used the old hosts to replace two nodes already in my ES cluster and add two more, for a total of five data nodes. That means I have one node with a CPU that benchmarks less than half the other four. Is this a problem? If so, would I actually get more speed out of the cluster by using the four faster nodes and removing the slower one?

---

<div class="post-metadata">

**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\
**Post date:** [March 10, 2020, 8:25pm UTC](https://discuss.elastic.co/t/affect-of-disparate-cpu-speeds-on-cluster/222962/2 "2020-03-10T20:25:02Z")

</div>

For the most part, querying Elasticsearch will go as fast as the slowest node.

---

<div class="post-metadata">

**Author:** ![HockeyFan0000](https://avatars.discourse-cdn.com/v4/letter/h/b4bc9f/32.png) [@HockeyFan0000](https://discuss.elastic.co/u/HockeyFan0000)\
**Post date:** [March 11, 2020, 11:45am UTC](https://discuss.elastic.co/t/affect-of-disparate-cpu-speeds-on-cluster/222962/3 "2020-03-11T11:45:06Z")

</div>

Thank you for your reply. I'm aware the search will only be as fast as the slowest node. My question is - will I be better off minus the slower node even though I'll have one fewer nodes.

---

<div class="post-metadata">

**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\
**Post date:** [March 11, 2020, 8:36pm UTC](https://discuss.elastic.co/t/affect-of-disparate-cpu-speeds-on-cluster/222962/4 "2020-03-11T20:36:09Z")

</div>

That depends on your requirements and how slow that node actually makes things.

You can always setup hot/warm/cold tiering to allow that node to hold data.

---

<div class="post-metadata">

**Author:** ![rugenl](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/rugenl/32/12887_2.png) [@rugenl](https://discuss.elastic.co/u/rugenl)\
**Post date:** [March 11, 2020, 11:16pm UTC](https://discuss.elastic.co/t/affect-of-disparate-cpu-speeds-on-cluster/222962/5 "2020-03-11T23:16:57Z")

</div>

You can use the Kibana profile tool with queries that will use both fast and slow nodes. See how much difference there is between them, then decide if it matters.

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\
**Post date:** [April 8, 2020, 11:16pm UTC](https://discuss.elastic.co/t/affect-of-disparate-cpu-speeds-on-cluster/222962/6 "2020-04-08T23:16:59Z")

</div>

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