# \>10K fields: is it a problem? Why exactly?

**URL:** <https://discuss.elastic.co/t/10k-fields-is-it-a-problem-why-exactly/359285>\
**Category:** Elasticsearch\
**Created:** [May 10, 2024, 9:53pm UTC](https://discuss.elastic.co/t/10k-fields-is-it-a-problem-why-exactly/359285 "2024-05-10T21:53:27Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Mikhail\_Khludnev](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/mikhail_khludnev/32/59591_2.png) [@Mikhail\_Khludnev](https://discuss.elastic.co/u/Mikhail_Khludnev)\
**Post date:** [May 10, 2024, 9:53pm UTC](https://discuss.elastic.co/t/10k-fields-is-it-a-problem-why-exactly/359285/1 "2024-05-10T21:53:27Z")

</div>

Hello.  
I have shards about 20G size with 600K docs each. Response time is acceptable.

After a shard receives about 200K updates, it ends up 20 segments, response time exceeds the reasonable value.

In profile, I spotted really high create\_weight time (300 ms) for CombinedFieldQuery (7 fields, 3 terms - not much), although underneath TermQueries create weight and scorers quite fast.

May many fields cause high create\_weight for CombinedFieldQuery?  
What are other usual implications? Which metrics/stats to check when number of fields is hight?

---

<div class="post-metadata">

**Author:** ![Mikhail\_Khludnev](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/mikhail_khludnev/32/59591_2.png) [@Mikhail\_Khludnev](https://discuss.elastic.co/u/Mikhail_Khludnev)\
**Post date:** [May 20, 2024, 7:40am UTC](https://discuss.elastic.co/t/10k-fields-is-it-a-problem-why-exactly/359285/2 "2024-05-20T07:40:37Z")

</div>

I've tried to reproduce it locally: built an 45G index with 10K fields in mapping and 1M docs, having about 30 segments. I don't see slow `create_weight` there.  
May you give a clue where to look at?