# Updated "hardware" recommendations?

**URL:** <https://discuss.elastic.co/t/updated-hardware-recommendations/290930>\
**Category:** Elasticsearch\
**Created:** [December 3, 2021, 8:34pm UTC](https://discuss.elastic.co/t/updated-hardware-recommendations/290930 "2021-12-03T20:34:51Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Jamshid](https://avatars.discourse-cdn.com/v4/letter/j/ecd19e/32.png) [@Jamshid](https://discuss.elastic.co/u/Jamshid)\
**Post date:** [December 3, 2021, 8:34pm UTC](https://discuss.elastic.co/t/updated-hardware-recommendations/290930/1 "2021-12-03T20:34:51Z")

</div>

Just wondering if there is an updated doc similar to the very helpful 2.3 guide

> **[Hardware | Elasticsearch: The Definitive Guide \[2.x\] | Elastic](https://www.elastic.co/guide/en/elasticsearch/guide/current/hardware.html)**

It summarized memory, disk, and network configuration information and is probably still relevant but I can't find a modern version of the page.

It still gets referenced in posts like:

> [@I/O scheduler for ES 7.0](https://discuss.elastic.co/t/i-o-scheduler-for-es-7-0/223981):
>
> We are using ES 7.5.0 As per the [documentation](https://www.elastic.co/guide/en/elasticsearch/guide/current/hardware.html#id-1.10.4.5.5.4.2.1), selecting the correct i/o scheduler should optimize the ES performance. In our system when I checked in /sys/block/sdb/queue/scheduler, I could see 3 scheduler options noop deadline cfq We are using azure blockstorage in our system. So which scheduler should be best for our configuration?

I guess it's spread now among pages like this:

> **[Tune for indexing speed | Elasticsearch Guide \[7.15\] | Elastic](https://www.elastic.co/guide/en/elasticsearch/reference/7.15/tune-for-indexing-speed.html#_use_faster_hardware)**

[https://www.elastic.co/guide/en/elasticsearch/reference/7.15/tune-for-search-speed.html](https://www.elastic.co/guide/en/elasticsearch/reference/7.15/tune-for-search-speed.html#_use_faster_hardware_2)

> **[Merge | Elasticsearch Guide \[7.15\] | Elastic](https://www.elastic.co/guide/en/elasticsearch/reference/7.15/index-modules-merge.html#merge-scheduling)**

> **[Important Elasticsearch configuration | Elasticsearch Guide \[7.10\] | Elastic](https://www.elastic.co/guide/en/elasticsearch/reference/7.10/important-settings.html#heap-size-settings)**

Maybe the I/O Scheduler info wrt SSDs is not relevant with modern Linux?  
Can't find the recommendation against cross-datacenter clusters.  
Also don't find current recommendations like "prefer medium-to-large boxes" or around number of CPU cores.

---

<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:** [December 31, 2021, 8:35pm UTC](https://discuss.elastic.co/t/updated-hardware-recommendations/290930/2 "2021-12-31T20:35:31Z")

</div>

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