# Self-hosted router - how to optimize Kubernetes and what is the limit?

**URL:** <https://community.apollographql.com/t/self-hosted-router-how-to-optimize-kubernetes-and-what-is-the-limit/7343>\
**Category:** Router\
**Tags:** federation\
**Created:** [March 29, 2024, 7:52am UTC](https://community.apollographql.com/t/self-hosted-router-how-to-optimize-kubernetes-and-what-is-the-limit/7343 "2024-03-29T07:52:59Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![billnbell](https://avatars.discourse-cdn.com/v4/letter/b/eb8c5e/32.png) [@billnbell](https://community.apollographql.com/u/billnbell)\
**Post date:** [March 29, 2024, 7:52am UTC](https://community.apollographql.com/t/self-hosted-router-how-to-optimize-kubernetes-and-what-is-the-limit/7343/1 "2024-03-29T07:52:59Z")

</div>

We are using router/supergraph and we were wondering how many super graphs we need for a high load event (like a Super Bowl ad). How many concurrent connections per machine? How have 0ther people scaled. I think # of sockets and TIME\_WAIT may be an issue. Unfortunately we are using EKS on Fargate and there does not appears to be a way to get networking sessions on Linux without privileged mode.?

Thanks