## Overview

In order to gather metrics on our system's ability to handle increased loads and users, we have performed tests to evaluate the maximum number of concurrent users a Rasa assistant can handle with certain machine configurations.
In each test case we spawned the following number of concurrent users at peak concurrency using a [spawn rate](https://docs.locust.io/en/1.5.0/configuration.html#all-available-configuration-options) of 1000 users per second.
In our tests we used the Rasa [HTTP-API](/content/docs/rasa/pages/http-api/index.html) and the [Locust](https://locust.io/) open source load testing tool.

| Users             | CPU                           | Memory  |
|-------------------|-------------------------------|---------|
| Up to 50,000     | 6vCPU                        | 16 GB   |
| Up to 80,000     | 6vCPU, with almost 90% CPU usage | 16 GB   |

### Debugging bot related issues while scaling up

To test the Rasa [HTTP-API](/content/docs/rasa/pages/http-api/index.html) ability to handle a large number of concurrent user activity we used the Rasa Pro [tracing](https://legacy-docs-oss.rasa.com/docs/rasa/next/monitoring/tracing) capability
along with a tracing backend or collector, such as Jaeger, to collect traces for the bot under test.

##### note

Our team is currently in the process of running additional performance-related tests. More information will be added here as we progress.