Nodies currently holds 20 app stakes staked into 5 chains, each 20M each per session (the other 6 we have are unused). Assuming best case scenario (or rather worst case scenario for block space) for our gateway, we take up 2MB of block space and have access to upwards to 9.6B relays a day assuming uniform distribution of traffic. (20M * 20 * 24). Obviously, the best case scenario is likely not to happen as our traffic is not uniform and not every node operator will be effective in a session. But it does give us a better general idea.
The other gateway operators that are coming onboard are expected to adopt the gateway server which will exhibit the same behaviors that Nodies gateway is showing on chain. With this, we should be able to scale upwards to roughly 3-4 more gateway (2MB each) operators if they have the same traffic patterns as us based off @RawthiL research
I support this proposal. Thanks everyone for the analysis and help