# Best approach to running my own indexer with low latency

**URL:** <https://forum.algorand.co/t/best-approach-to-running-my-own-indexer-with-low-latency/5612>\
**Category:** General\
**Created:** [January 3, 2022, 11:52pm UTC](https://forum.algorand.co/t/best-approach-to-running-my-own-indexer-with-low-latency/5612 "2022-01-03T23:52:20Z")\
**Posts on this page:** 1\
**Showing post:** 2

<div class="post-metadata">

**Author:** ![fabrice](https://avatars.discourse-cdn.com/v4/letter/f/5fc32e/32.png) [@fabrice](https://forum.algorand.co/u/fabrice)\
**Post date:** [January 5, 2022, 3:34pm UTC](https://forum.algorand.co/t/best-approach-to-running-my-own-indexer-with-low-latency/5612/2 "2022-01-05T15:34:12Z")

</div>

It depends on a lot of parameters.  
I think that in general, the cloud is easier, requires less capital (no need to buy servers), and less maintenance (if the server crashes, the VM is moved easily for example).  
On the other hand, it is more expensive on the long term from a purely hardware cost metric.

From my experience, what is the most expensive running an archival algod + indexer is the storage: you need close to 2TB SSD nowadays if you want low latency. And this keeps increasing:

> **[How Big Is Algorand?](https://howbigisalgorand.com/)**
>
> How big is an Algorand archival node?

---

_[View the full topic](https://forum.algorand.co/t/best-approach-to-running-my-own-indexer-with-low-latency/5612)._