# Multi-Tier Architecture, best configuration for Elasticsearch.output **URL:** \ **Category:** Elasticsearch\ **Tags:** ilm-index-lifecycle-management, elastic-agent\ **Created:** [July 18, 2022, 9:04am UTC](https://discuss.elastic.co/t/multi-tier-architecture-best-configuration-for-elasticsearch-output/309862 "2022-07-18T09:04:01Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![A\_Abdellah](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/a_abdellah/32/100734_2.png) [@A\_Abdellah](https://discuss.elastic.co/u/A_Abdellah)\ **Post date:** [July 18, 2022, 9:04am UTC](https://discuss.elastic.co/t/multi-tier-architecture-best-configuration-for-elasticsearch-output/309862/1 "2022-07-18T09:04:01Z")
Hello, I hope you are having a good day, I'm trying to migrate to a multi-tier architecture, and I'm confused what should the `Elasticsearch.output` configuration in beats and Elastic Agent be: - Should I configure beats to output only to hot data nodes ? - Or should I use all nodes for better ingestion and wait for allocation to move indices to hot nodes ? ---
**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:** [July 19, 2022, 1:25am UTC](https://discuss.elastic.co/t/multi-tier-architecture-best-configuration-for-elasticsearch-output/309862/2 "2022-07-19T01:25:23Z")
If you have hot nodes then that is likely where the data is being indexed, and not on any other nodes, so use them. ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [August 16, 2022, 1:25am UTC](https://discuss.elastic.co/t/multi-tier-architecture-best-configuration-for-elasticsearch-output/309862/3 "2022-08-16T01:25:37Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.