# Sizing question **URL:** \ **Category:** Elasticsearch\ **Created:** [May 5, 2022, 9:56am UTC](https://discuss.elastic.co/t/sizing-question/304014 "2022-05-05T09:56:30Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![marek\_duciuc](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/marek_duciuc/32/100189_2.png) [@marek\_duciuc](https://discuss.elastic.co/u/marek_duciuc)\ **Post date:** [May 5, 2022, 9:56am UTC](https://discuss.elastic.co/t/sizing-question/304014/1 "2022-05-05T09:56:30Z")
Elasticsearch recommendations is as follows: Maximum of 32 GB heap 20 shards per 1 GB of heap 40 to 60 GB of data per shard this gives approximately 25TB to 39TB of data per node. Is this only theoretical or would it really work? ---
**Author:** ![Christian\_Dahlqvist](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/christian_dahlqvist/32/4617_2.png) [@Christian\_Dahlqvist](https://discuss.elastic.co/u/Christian_Dahlqvist)\ **Post date:** [May 5, 2022, 11:01am UTC](https://discuss.elastic.co/t/sizing-question/304014/2 "2022-05-05T11:01:09Z")
> [@marek\_duciuc](#): > > 20 shards per 1 GB of heap This is a maximum, not a recommended target. The number will depend on shard size, mappings, index settings and node load. Note that this recommendation was created for an older version of Elasticsearch and may have changed since. > [@marek\_duciuc](#): > > Is this only theoretical or would it really work? This depends on what your requirements are. Improvements in recent versions have reduced the amount of memory required per shard, but having lots of data on a node naturally affect query performance. ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [June 2, 2022, 11:01am UTC](https://discuss.elastic.co/t/sizing-question/304014/3 "2022-06-02T11:01:29Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.