# Deployment of Rasa ,1000 concurrent user at a time

**URL:** <https://forum.rasa.com/t/deployment-of-rasa-1000-concurrent-user-at-a-time/59308>\
**Category:** Rasa Open Source\
**Created:** [August 11, 2023, 3:33am UTC](https://forum.rasa.com/t/deployment-of-rasa-1000-concurrent-user-at-a-time/59308 "2023-08-11T03:33:09Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Sanju7820](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/sanju7820/32/23644_2.png) [@Sanju7820](https://forum.rasa.com/u/Sanju7820)\
**Post date:** [August 11, 2023, 3:33am UTC](https://forum.rasa.com/t/deployment-of-rasa-1000-concurrent-user-at-a-time/59308/1 "2023-08-11T03:33:09Z")

</div>

What are the basic hardware required and how to deploy rasa in server in which rasa can tackle around 1000 concurrent user at a time

---

<div class="post-metadata">

**Author:** ![stephens](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/stephens/32/1632_2.png) [@stephens](https://forum.rasa.com/u/stephens)\
**Post date:** [August 11, 2023, 3:15pm UTC](https://forum.rasa.com/t/deployment-of-rasa-1000-concurrent-user-at-a-time/59308/2 "2023-08-11T15:15:49Z")

</div>

There are several factors in determing this including your model pipeline and deployment architecture and dependencies. You should do load testing and use Kubernetes for a scalable deployment. It’s best to assume you will need multiple instances and want to use Kubernetes to scale the number of instances based on demand.

There’s a brief discussion of load testing [here](https://rasa.com/docs/rasa/monitoring/load-testing-guidelines/).

Rasa provides services and tools to help with this if you are under contract. You can contact sales for more information on this.