# Separate index and search workloads without remote-backed storage

**URL:** <https://forum.opensearch.org/t/separate-index-and-search-workloads-without-remote-backed-storage/28047>\
**Category:** OpenSearch\
**Tags:** configure\
**Created:** [May 5, 2026, 1:14pm UTC](https://forum.opensearch.org/t/separate-index-and-search-workloads-without-remote-backed-storage/28047 "2026-05-05T13:14:35Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![vit](https://avatars.discourse-cdn.com/v4/letter/v/65b543/32.png) [@vit](https://forum.opensearch.org/u/vit)\
**Post date:** [May 5, 2026, 1:14pm UTC](https://forum.opensearch.org/t/separate-index-and-search-workloads-without-remote-backed-storage/28047/1 "2026-05-05T13:14:35Z")

</div>

**Versions** : 3.4

**Describe the issue** : The documentation states that separate indexing and search tasks are possible in a cluster with remote storage enabled. Remote-backed storage use for segment replication. But segment replication also support node-to-node communication. Is it possible to allocate search nodes in a cluster without using remote storage?

**Configuration** :

**Relevant Logs or Screenshots** :

---

<div class="post-metadata">

**Author:** ![pablo](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.opensearch.org/pablo/32/10363_2.png) [@pablo](https://forum.opensearch.org/u/pablo)\
**Post date:** [May 5, 2026, 1:16pm UTC](https://forum.opensearch.org/t/separate-index-and-search-workloads-without-remote-backed-storage/28047/2 "2026-05-05T13:16:00Z")

</div>

@vit According to OpenSearch documentation, the search node role is designed for remote-store-enabled clusters with segment-replication-enabled indexes.

> **[Separate index and search workloads](https://docs.opensearch.org/latest/tuning-your-cluster/separate-index-and-search-workloads/)**
>
> Separate index and search workloads

---

<div class="post-metadata">

**Author:** ![vit](https://avatars.discourse-cdn.com/v4/letter/v/65b543/32.png) [@vit](https://forum.opensearch.org/u/vit)\
**Post date:** [May 5, 2026, 1:23pm UTC](https://forum.opensearch.org/t/separate-index-and-search-workloads-without-remote-backed-storage/28047/3 "2026-05-05T13:23:23Z")

</div>

Yes, I read that section of the documentation. But is this a requirement? As far as I understand, remote storage is only needed for replication. Perhaps a second configuration option is possible, enabling node-to-node replication?