Content
@
0 reply
21 recasts
21 reactions
Varun Srinivasan
@v
We're scaling hubs to handle 10x traffic (~ 500k DAU) @adityapk @sanjay and @wazzymandias.eth are heads down and making some big changes this month. Here's a quick overview of how hubs work today.
9 replies
47 recasts
244 reactions
Varun Srinivasan
@v
An important part of hub performance is catch-up sync or diff sync, which kicks in when hubs get disconnected for a short duration. @sanjay developed this back in 2022 when we were first designing hubs.
2 replies
0 recast
9 reactions
Sina Habibian
@sinahab
It all makes sense!
0 reply
0 recast
2 reactions
kevin j 🤗
@entropybender
can we get a vid after changes comparing the prev design with post? would be so cool to see, esp if there are benchmarks
0 reply
0 recast
1 reaction
The Old Castle Defense(TOCD)
@tocd
Great to see your transparency FC team. It's an amazing example how team should do any business. 🚀lfg
0 reply
0 recast
0 reaction
daws.eth 🎩 @farcon
@daws
🫡 Fan of that TypeScript + Rust combo
0 reply
0 recast
0 reaction
The Wizard 🎩
@queue
We love a team that builds in public Excited to see these changes go live 🤝
0 reply
0 recast
0 reaction
⚪️ cooper 🎩 ↑
@cooperray
this is the good stuff. crucial upgrades. thanks for sharing the detail 1069 $degen
0 reply
0 recast
0 reaction
Ed 🎩
@dralle
Great explainer & fantastic work 😎 369 $DEGEN
0 reply
0 recast
0 reaction
Chase
@harden-hardys
I know you and Dan are familiar with Urbit, so I'm curious why you didn't choose a more heavily sharded Hub design from the outset? Urbit shards the network by channel (called a Group). A group host is generally a dedicated node, tho some people use their daily driver. Despite being WAY less performant than Hubs...
1 reply
0 recast
0 reaction
