It's time to replace TCP in the datacenter (2023)
1–10 of 160 posts
Re: It's time to replace TCP in the datacenter (2023)
#2Previous discussions:
Homa, a transport protocol to replace TCP for low-latency RPC in data centers https://news.ycombinator.com/item?id=28204808
Linux implementation of Homa https://news.ycombinator.com/item?id=28440542
Re: It's time to replace TCP in the datacenter (2023)
#3The original paper was discussed previously at https://news.ycombinator.com/item?id=33401480
Re: It's time to replace TCP in the datacenter (2023)
#4Homa: A Receiver-Driven Low-Latency Transport Protocol Using Network Priorities
https://people.csail.mit.edu/alizadeh/papers/homa-sigcomm18....
Re: It's time to replace TCP in the datacenter (2023)
#5TCP was replaced in the data centers of certain FAANG companies years before this paper.
Re: It's time to replace TCP in the datacenter (2023)
#6TCP was replaced in the data centers of certain FAANG companies years before this paper.
If they keep it secret they don't get credit for it.
Re: It's time to replace TCP in the datacenter (2023)
#7TCP was replaced in the data centers of certain FAANG companies years before this paper.
*minority of the fangs.
Re: It's time to replace TCP in the datacenter (2023)
#8> If Homa becomes widely deployed, I hypothesize that core congestion will cease to exist as a significant networking problem, as long as the core is not systemically overloaded.
Yep. Sure; but, what happens when it becomes overloaded?
> Homa manages congestion from the receiver, not the sender. [...] but the remaining scheduled packets may only be sent in response to grants from the receiver
I hypothesize it will not be a great day when you do become "systemically" overloaded.
Re: It's time to replace TCP in the datacenter (2023)
#9Another thing not worth investing time into for the rest of our careers. TCP will be around for decades to come.