# Used split API, resulting index is much, much larger than the source index **URL:** \ **Category:** Elasticsearch\ **Created:** [June 4, 2024, 2:05pm UTC](https://discuss.elastic.co/t/used-split-api-resulting-index-is-much-much-larger-than-the-source-index/360792 "2024-06-04T14:05:31Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![daviddawson](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/daviddawson/32/135024_2.png) [@daviddawson](https://discuss.elastic.co/u/daviddawson)\ **Post date:** [June 4, 2024, 2:05pm UTC](https://discuss.elastic.co/t/used-split-api-resulting-index-is-much-much-larger-than-the-source-index/360792/1 "2024-06-04T14:05:31Z")
An index in our cluster was created by accident with a single shard. We looked at various ways to expand the shards, as performance is suffering. It's currently about 280gb. We tried the split API to generate a new index from it with 20 shards. The resulting index took a while to rebalance the shards across the cluster, but the size of the index seems to be very large, and not reducing. It's at 5tb now, which make sense for the way split works. The issue is that it's been sat at 5tb for a day now, and doesn't seem to be reducing down as I would have expected. Any hints on what I should be looking for or expecting with the split api used like this? ---
**Author:** ![leandrojmp](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/leandrojmp/32/107231_2.png) [@leandrojmp](https://discuss.elastic.co/u/leandrojmp)\ **Post date:** [June 4, 2024, 3:00pm UTC](https://discuss.elastic.co/t/used-split-api-resulting-index-is-much-much-larger-than-the-source-index/360792/2 "2024-06-04T15:00:24Z")
> [@daviddawson](#): > > We tried the split API to generate a new index from it with 20 shards. How many replicas do you have? Please share the response of `GET _cat/shards` running on Kibana Dev Tools.