We need a replacement for TCP in the datacenter [pdf]

Out of order delivery is fine in TCP within the window. It might be inefficient but it's not impossible, reassembly could be moved to userspace if userspace TCP was used.

I have no problem with alternates to TCP in the DC with a crossbar fabric and far less loss, seems sensible.

I wonder how it would play with QUIC and the session like behaviours now emerging.

Re: We need a replacement for TCP in the datacenter [pdf]

#5

Post by defrost » Mon, Oct 31, 2022, 3:06 AM UTC

Jumping to the end:

> TCP is the wrong protocol for datacenter computing.

> Every aspect of TCP’s design is wrong: there is no part worth keeping.

I cannot disagree and Ousterhout argues well.

> Homa offers an alternative that appears to solve all of TCP’s problems.

I'm well behind the curve on protocols and now I have something to learn more about.

> The best way to bring Homa into widespread usage is integrate it with the RPC frameworks that underly most large-scale datacenter applications.

More or less the case for whatever replaces TCP in a tight computing warehouse setup.

Re: We need a replacement for TCP in the datacenter [pdf]

#6

Post by kevin_thibedeau » Mon, Oct 31, 2022, 3:08 AM UTC

If only we had a stream control transport protocol and were allowed to use it.

Re: We need a replacement for TCP in the datacenter [pdf]

#7

Post by manv1 » Mon, Oct 31, 2022, 3:09 AM UTC

It's interesting that everyone (including the author) talks about UDP as a lossy protocol, but it doesn't seem that UDP drops actually occur on a routine basis anywhere. The UDP-based DDOS attacks seem to prove that; if UDP really was being dropped, those DDOS attacks wouldn't be so problematic.

That said, it's an interesting read. TCP is inefficient, but that inefficiency has been patched/masked by hardware solutions.

It's nice to see that someone's still thinking about this this. I remember the days when there were tons of non-IP protocols floating around (IPX, DecNET, AppleTalk, etc). TCP/IP won, which was not an obvious thing at the time.

Re: We need a replacement for TCP in the datacenter [pdf]

#8

Post by vore » Mon, Oct 31, 2022, 3:14 AM UTC

manv1 wrote: post #7

It's interesting that everyone (including the author) talks about UDP as a lossy protocol, but it doesn't seem that UDP drops actually occur on a routine basis anywhere. The UDP-based DDOS attacks seem to prove that; if UDP really was being dropped, those DDOS attacks wouldn't be so problematic. That said, it's an interesting read. TCP is inefficient, but that inefficiency has been patched/masked by hardware solution…

I think re UDP the point regarding it being unreliable is that you have to design your applications to take the unreliability into account, because it does happen even if it may be infrequent: assuming that it is reliable when you can get unreliable behavior will result in correctness issues.

Re: We need a replacement for TCP in the datacenter [pdf]

#9

Post by annoyingnoob » Mon, Oct 31, 2022, 3:14 AM UTC

Homa https://dl.acm.org/doi/abs/10.1145/3230543.3230564

Re: We need a replacement for TCP in the datacenter [pdf]

#10

Post by BirAdam » Mon, Oct 31, 2022, 3:16 AM UTC

Well, the key would be to develop and deploy Homa in a DC and test in implementation at scale. If it actually ameliorates the perceived shortcomings of TCP that make nothing in TCP worth keeping as this author says, then cool. My only complain with issues like this is the cost of implementation. Someone has to pay to build a DC around it or increase the cost of maintenance for several years to a decade while supporting two completely different and incompatible networks.