It's Time to Replace TCP in the Datacenter

TL;DR

This paper argues that TCP is unsuitable for modern datacenters and introduces Homa, a new transport protocol designed to overcome TCP's fundamental limitations, promising better performance and integration with existing frameworks.

Contribution

The paper presents Homa, a novel transport protocol tailored for datacenter environments, addressing TCP's core issues and enabling improved network performance.

Findings

Homa avoids TCP's problems like in-order delivery and stream orientation.

Homa can be integrated with RPC frameworks for widespread adoption.

Homa demonstrates superior performance in datacenter scenarios.

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.