# Hardware requirements for high performant search on a 200M vector dataset

**URL:** https://discuss.elastic.co/t/hardware-requirements-for-high-performant-search-on-a-200m-vector-dataset/356340
**Category:** Elasticsearch
**Tags:** elastic-stack-machine-learning
**Created:** [March 27, 2024, 9:23pm UTC](https://discuss.elastic.co/t/hardware-requirements-for-high-performant-search-on-a-200m-vector-dataset/356340 "2024-03-27T21:23:13Z")
**Posts on this page:** 3
**Page:** 1

<div class="post-metadata">

### Author: ![nadasaiyed](https://avatars.discourse-cdn.com/v4/letter/n/45deac/32.png) [@nadasaiyed](https://discuss.elastic.co/u/nadasaiyed)
#### Post date: [March 27, 2024, 9:23pm UTC](https://discuss.elastic.co/t/hardware-requirements-for-high-performant-search-on-a-200m-vector-dataset/356340/1 "2024-03-27T21:23:13Z")

</div>

I am trying to understand how much hardware will I need to store 50M vectors of dimension 1536, projected to reach 200M. I have read HNSW index is stored in memory, will I need as much RAM as the data to do a similarity search?

---

<div class="post-metadata">

### Author: ![asad-awadia](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/asad-awadia/32/133104_2.png) [@asad-awadia](https://discuss.elastic.co/u/asad-awadia)
#### Post date: [March 29, 2024, 1:09am UTC](https://discuss.elastic.co/t/hardware-requirements-for-high-performant-search-on-a-200m-vector-dataset/356340/2 "2024-03-29T01:09:54Z")

</div>

Interested in knowing this as well - want to know how to size my cluster

---

<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 26, 2024, 1:10am UTC](https://discuss.elastic.co/t/hardware-requirements-for-high-performant-search-on-a-200m-vector-dataset/356340/3 "2024-04-26T01:10:27Z")

</div>

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