# Capacity planning tools?

**URL:** <https://discuss.elastic.co/t/capacity-planning-tools/25115>\
**Category:** Elasticsearch\
**Created:** [July 8, 2015, 6:01am UTC](https://discuss.elastic.co/t/capacity-planning-tools/25115 "2015-07-08T06:01:39Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![linlma](https://avatars.discourse-cdn.com/v4/letter/l/ec9cab/32.png) [@linlma](https://discuss.elastic.co/u/linlma)\
**Post date:** [July 8, 2015, 6:01am UTC](https://discuss.elastic.co/t/capacity-planning-tools/25115/1 "2015-07-08T06:01:39Z")

</div>

Hello Elastic experts,

Is there any capacity planning tools? For example, given # of records to index and use cases (with SLA), how much H/W is needed?

thanks in advance,  
Lin

---

<div class="post-metadata">

**Author:** ![magnusbaeck](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/magnusbaeck/32/44943_2.png) [@magnusbaeck](https://discuss.elastic.co/u/magnusbaeck)\
**Post date:** [July 8, 2015, 7:03am UTC](https://discuss.elastic.co/t/capacity-planning-tools/25115/2 "2015-07-08T07:03:26Z")

</div>

This is extremely hard to do since there are too many factors. Ignoring the variety in hardware at least the following factors contribute:

- Number of indexes.
- Number of nodes.
- Number of shards.
- Number of documents per index.
- Frequency of write operations.
- Frequency of deletions.
- Size of each document.
- How documents are mapped.
- Frequency of queries.
- Types of queries.
- Whether indexes are queried uniformly or if there are "hot spots".

---

<div class="post-metadata">

**Author:** ![linlma](https://avatars.discourse-cdn.com/v4/letter/l/ec9cab/32.png) [@linlma](https://discuss.elastic.co/u/linlma)\
**Post date:** [July 8, 2015, 11:32pm UTC](https://discuss.elastic.co/t/capacity-planning-tools/25115/3 "2015-07-08T23:32:11Z")

</div>

Thanks Magnus,

If so, what is your suggested way of doing capacity planning? Doing real testing?

BTW, what do you mean "documents are mapped"?

regards,  
Lin

---

<div class="post-metadata">

**Author:** ![magnusbaeck](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/magnusbaeck/32/44943_2.png) [@magnusbaeck](https://discuss.elastic.co/u/magnusbaeck)\
**Post date:** [July 9, 2015, 3:45am UTC](https://discuss.elastic.co/t/capacity-planning-tools/25115/4 "2015-07-09T03:45:17Z")

</div>

Yes, I advocate real-world testing and being prepared to scale your cluster.

By "How documents are mapped" I mean how Elasticsearch analyzes (tokenizes) and indexes a text. This affects how many tokens (terms) are added to the index and together with the type of queries made it plays an important part of the cluster's query performance.

---

<div class="post-metadata">

**Author:** ![linlma](https://avatars.discourse-cdn.com/v4/letter/l/ec9cab/32.png) [@linlma](https://discuss.elastic.co/u/linlma)\
**Post date:** [July 9, 2015, 4:02am UTC](https://discuss.elastic.co/t/capacity-planning-tools/25115/5 "2015-07-09T04:02:55Z")

</div>

Thanks Magnus!

regards,  
Lin

---

<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:** [July 6, 2017, 12:02am UTC](https://discuss.elastic.co/t/capacity-planning-tools/25115/6 "2017-07-06T00:02:40Z")

</div>