# Dense vector disk size

**URL:** <https://discuss.elastic.co/t/dense-vector-disk-size/365705>\
**Category:** Elasticsearch\
**Tags:** vector-search\
**Created:** [August 28, 2024, 3:05pm UTC](https://discuss.elastic.co/t/dense-vector-disk-size/365705 "2024-08-28T15:05:29Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![amberry](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/amberry/32/137121_2.png) [@amberry](https://discuss.elastic.co/u/amberry)\
**Post date:** [August 28, 2024, 3:05pm UTC](https://discuss.elastic.co/t/dense-vector-disk-size/365705/1 "2024-08-28T15:05:29Z")

</div>

Hi there,

We're attempting to index about 10,000,000 `1536`-dim vectors. Per the documentation, we've calculated this should take up `10M*(4*1536+4)` = approximately 61.5GB. In reality, this index is taking up about 250GB. (There are several other keyword/text fields, a mix of indexed/unindexed, but we don't believe these will be the reason the size is exploding?)

What are we doing wrong/where have we gone wrong in our calculations? Alternatively, is there anything we can do to reduce this disk size?

We also tried to set `index_type` to `int4_hnsw` , which unexpectedly led to a larger rather than smaller size.  
For reference, we are on version `8.15.0`.

---

<div class="post-metadata">

**Author:** ![Kathleen\_DeRusso](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/kathleen_derusso/32/132039_2.png) [@Kathleen\_DeRusso](https://discuss.elastic.co/u/Kathleen_DeRusso)\
**Post date:** [August 28, 2024, 6:34pm UTC](https://discuss.elastic.co/t/dense-vector-disk-size/365705/2 "2024-08-28T18:34:01Z")

</div>

Welcome @amberry !

Off the cuff, your vector math seems reasonable but it's hard to say how much impact your other fields would have.

Here's what I would check:

- Do you have a lot of unmerged segments with deleted documents that could be inflating your disk size?
- You can see if anything glaring comes out of the [Disk usage API](https://www.elastic.co/guide/en/elasticsearch/reference/current/indices-disk-usage.html)
- You could look at some other suggestions for [tuning kNN search](https://www.elastic.co/guide/en/elasticsearch/reference/current/tune-knn-search.html) or just our [general recommendations for tuning for disk usage](https://www.elastic.co/guide/en/elasticsearch/reference/current/tune-for-disk-usage.html)