Skip to content

Conversation

karankurbur
Copy link

@karankurbur karankurbur commented Jul 15, 2025

Extends the current rollup-boost HA spec to handle a large number of client requests.

Copy link

vercel bot commented Jul 15, 2025

The latest updates on your projects. Learn more about Vercel for Git ↗︎

1 Skipped Deployment
Name Status Preview Comments Updated (UTC)
rollup-boost ⬜️ Ignored (Inspect) Visit Preview Jul 15, 2025 5:29am

@karankurbur karankurbur marked this pull request as draft July 15, 2025 05:53

The following design introduces two components to the existing HA setup:

A queue that will replace websockets for communcation between `rollup-boost` <-> `reverse-proxy` The current approach has a single `websocket-proxy` that is a single point of failure. As the number of client connections increase, scaling up the networking/compute of the single instance also becomes a limitation. Without a message queue, each rollup-boost instance would also need to manage connections to many `reverse-proxy` as they may be scaled up/down.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The current approach has a single websocket-proxy that is a single point of failure

FYI you can run multiple websocket-proxy's connected to each rollup-boost instance. What we have is:

5 node HA sequencer cluster <-- N websocket proxy cluster <--- ALB

This has scaled up to a fairly large number of external clients, as each websocket proxy can handle quite a large number

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants