Hacker News Reader

So they're re-inventing HPC networks in the data center.

https://en.wikipedia.org/wiki/Fat_tree

https://www.cs.umd.edu/class/spring2021/cmsc714/readings/Kim...

I'm sure there are innovations here, but most of this has been standard in HPC for decades. (Fat trees since 1985, Dragonfly since 2008.) This is not new science, folks.

View on HN

It's not new science, but tuning RoCE performance is new engineering.

And honestly the difference between science and engineering in systems is overrated. The real difference is "systems that work, or everyone knows how to run" vs "systems that are a bit of black magic".

While on paper it is true that HPC has had collectives and dragonflies and RDMA for ages, a whole host of issues make tuning these things hard and tedious. Such work also informs the design of future "smart networks" by keeping everyone updated with common problems and workarounds that folks employ.

Reply on news.ycombinator.com