HackerPapers

arXiv:2210.00714cs.NI

It's Time to Replace TCP in the Datacenter

John Ousterhout

Abstract

In spite of its long and successful history, TCP is a poor transport protocol for modern datacenters. Every significant element of TCP, from its stream orientation to its expectation of in-order packet delivery, is wrong for the datacenter. It is time to recognize that TCP's problems are too fundamental and interrelated to be fixed; the only way to harness the full performance potential of modern networks is to introduce a new transport protocol into the datacenter. Homa demonstrates that it is possible to create a transport protocol that avoids all of TCP's problems. Although Homa is not API-compatible with TCP, it should be possible to bring it into widespread usage by integrating it with RPC frameworks.

arXiv comments
Journal reference
DOI
Subjects
cs.NINetworking and Internet Architecture

On Hacker News

Linked by 3 identifiable commenters in 5 comments across 4 threads, from Nov 2022 to Oct 2023. Found in 2 HN submissions from Oct 2022 to Nov 2024. Open submission discussions.

Comments linking this paper, per year.

Discussed in comments4 threads

  1. A small warning about UDP based protocols 1 comment

    • Yes. Homa for one: * (2018) https://people.csail.mit.edu/alizadeh/papers/homa-sigcomm18.... * (2022) https://arxiv.org/abs/2210.00714 * (2022) https://www.theregister.com/2022/07/27/replace_tcp_datacente... * https://github.com/PlatformLab/Homa

      defrostView on HNParent context

  2. Unbounded memory usage by Linux TCP for receive buffers, and how we fixed it 1 comment

    • Great insights, and I would add that it’s not just the internet consumer side where there is pressure on TCP evolution, it’s in the datacenter, too. [1] [1] https://arxiv.org/abs/2210.00714

      tedd4uView on HNParent context

  3. Is it time to replace TCP in data centers? 2 comments

    • This seems like a review of John Ousterhout's work w/ Homa. I highly recommend reading the original. https://arxiv.org/abs/2210.00714 For those who don't know Ousterhout created the first log structured filesystem and created TCL (used heavily in hardware verification but also…

      shanemhansenView on HNParent context

    • QUIC could be a candidate, RPC over UDP/IP is definitely the direction proposed by Ousterhout. He has a few specific TCP things he's listed as being undesirable and frankly I'm not sure how QUIC stacks up on. - Bandwidth sharing (“fair” scheduling) - Sender-driven congestion…

      shanemhansenView on HNParent context

  4. Moving past TCP in the data center, part 2 1 comment

    • This short paper answers this question and more. I found it a helpful explanation of specifically what TCP design problems Homa is intended to address. https://arxiv.org/abs/2210.00714

      tedd4uView on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.