Skip to main content

Transport Options

Setup first: if you haven't read Why AI Networks Are Different, start there — it covers what transport is and why AI breaks the everyday baseline.

This page is the design space. Four distinct families, each solving a different problem, all coexisting in 2026. The history of how we got here fits in three eras.

The transport landscape, 2x2 grid of families. Family 1 Classic IP (L4) — general networking, kernel-bound — TCP, UDP, QUIC, MPTCP. Family 2 RDMA traditional — kernel-bypass HPC/AI, needs lossless underlay — InfiniBand, RoCE v2, RoCE v1, iWARP. Family 3 Hyperscaler custom — 100K+ GPU scale, drops PFC, packet spraying, microsecond failover — MRC, Falcon, SRD, UET. Family 4 Scale-up — intra-server GPU, TB/s, ns latency — NVLink, UALink, SUE, ICI.
The transport landscape at a glance. Each family solves a different problem and they coexist.

50 years in three eras

Transport evolutionTCP/IP1973UDP1980SCTP2000QUIC2012MPTCP2013Myrinet/Quadrics1995–2010InfiniBand1999iWARP2007RoCE v12010–2014RoCE v22014Pony Express2014–2023AWS EFA / SRD2018Alibaba eRDMA2020Spectrum-X2023UEC2023Google Falcon2023MRC2024197019801990200020102015202020261973 · TCP/IP1999 · IBTA formed2010 · RoCE v12012 · QUIC (Google)2014 · RoCE v22015 · DCQCN paper2019 · NVIDIA buys Mellanox2023 · UEC forms2024 · MRC (OCP)2025 · UEC 1.0
Active spec / standardVendor / proprietarySuperseded

Era 1 — TCP rules everything (1973–2000)

TCP/IP was designed for unreliable WANs in the 1970s. Reliability over latency. The world adopted it; everything else (web, mail, RPC) built on top. UDP (1980) and SCTP (2000) filled niches but TCP carried the load. By 2000, the network was a kernel-bound byte-stream with software-only congestion control — fine for everything except low-latency HPC.

Era 2 — HPC and early AI demand kernel-bypass (1999–2015)

In 1999 the IBTA defined RDMA and the verbs API, and InfiniBand shipped with credit-based flow control — bytes moved without the kernel touching them. RoCE v1 (2010) brought RDMA to Ethernet by faking a lossless fabric with PFC; RoCE v2 (2014) made it routable over UDP/IP. The DCQCN paper (Microsoft Research, 2015) proved RoCE could run at hyperscale, and RoCE moved from research into Azure, Meta, Tencent, and most cloud RDMA deployments today.

Era 3 — AI at 100K+ GPUs breaks RoCEv2 (2015 → today)

At 100K-GPU scale, RoCEv2's two crutches both fail. PFC causes head-of-line blocking and the deadly PFC storm; ECMP can't load-balance the elephant flows of a synchronized collective. So each hyperscaler built their own transport that drops both crutches: AWS SRD (2018), Alibaba eRDMA (2020), Google Falcon (2023), and MRC (2024 — the OpenAI / Microsoft / NVIDIA / AMD / Broadcom / Intel collaboration). The Ultra Ethernet Consortium (2023, ~50 members without NVIDIA) shipped UEC 1.0 in 2025 as the open-standard convergence target. NVIDIA, meanwhile, ships Spectrum-X — its own vertically-integrated AI-Ethernet stack.

The network is no longer just packet-forwarding infrastructure — it is part of the distributed compute system itself.


The four families

The transport landscape today sits in four buckets. Each solves a different problem, and they coexist:

#FamilySolves
1Classic IP transportsGeneral networking — internet, applications, control plane
2RDMA transports (traditional)Kernel-bypass for HPC and traditional AI clusters
3AI / hyperscaler custom transportsRoCEv2 at 100K+ GPU scale (multipath, no PFC dependence)
4Scale-up interconnectsIntra-server / intra-rack GPU-to-GPU communication

Pick the family you care about — each tab is self-contained:

You know these. Listed for completeness.

ProtocolOwner / StdReliableOrderedMultipathEncryptionKey traitUsed by / for
TCPIETFYesYesNoExternal (TLS)Byte-stream, AIMD CC, HoL blockingWeb, SSH, SMTP — universal
UDPIETFNoNoNoExternal (DTLS)Connectionless, low overheadDNS, DHCP, VoIP, gaming, QUIC base
SCTPIETFYesPer-streamMulti-home failoverExternal (DTLS)Multi-streaming + multi-homingTelecom — SS7/SIGTRAN, Diameter, 5G N2
DCCPIETFNoNoNoNoUnreliable + congestion controlMostly research / abandoned
QUICIETF (Google origin)YesPer-streamConnection migrationBuilt-in (TLS 1.3)0/1-RTT setup, user-spaceHTTP/3 — Google, Cloudflare, Meta, Apple
MPTCPIETFYesYesYes (subflows)ExternalTCP across multiple pathsApple Siri/iOS, Samsung, Linux
UDP-LiteIETFNoNoNoExternalPartial checksumLoss-tolerant codecs

Standout: QUIC. The only L4 protocol that natively supports multipath (via connection migration), is fully user-space, and ships with TLS 1.3 built in. You'll meet it on the inference and edge path.


Mental model — the 6-point synthesis

  1. Classic IP transports cover the internet. TCP, UDP, QUIC. Universal but kernel-bound.
  2. RDMA family (IB, RoCE v2, iWARP) covers traditional HPC/AI — but needs a lossless fabric (PFC) and doesn't multipath well.
  3. Each hyperscaler built a custom multipath transport because RoCEv2 doesn't scale to 100K+ GPUs: Google → Falcon, AWS → SRD, OpenAI/Microsoft/NVIDIA/AMD → MRC, Alibaba → eRDMA.
  4. UET is the open-standard convergence target. Expect MRC and Falcon ideas to fold into it over time.
  5. Scale-up (NVLink / UALink / SUE / ICI) is intra-server and disjoint from scale-out transports. Different domain, different physics, different protocols.
  6. Congestion control matters as much as the transport. Most modern AI fabrics combine packet spraying + delay-based CC + ECN/INT signals + microsecond failover.

Who built what — full reference table
TechOwner / Standards bodyWhen
TCP/IPIETF (DARPA)1973–1980s
UDPIETF1980
InfiniBand specIBTA consortium1999
Mellanox IB siliconMellanox Technologies (Israel)1999 → acquired by NVIDIA 2019
RDMA Verbs APIIBTA / OpenFabrics Alliance2000s
SCTPIETF2000
RoCE v1IBTA2010
QUICGoogle → IETF2012 / RFC 9000 in 2021
MPTCPIETF2013
RoCE v2IBTA2014
DCQCNMicrosoft ResearchSIGCOMM 2015
Pony ExpressGoogle (legacy)~2014–2023
AWS EFA / SRDAWS2018+
eRDMAAlibaba2020
Spectrum-XNVIDIA2023
Ultra Ethernet ConsortiumAMD, Arista, Broadcom, Cisco, HPE, Intel, Meta, Microsoft, +50 othersFounded 2023, spec 1.0 in 2025
FalconGoogle + Intel (E2100 IPU)2023
MRCOpenAI + AMD + Microsoft + NVIDIA + Broadcom + Intel (OCP)2024
UALink ConsortiumAMD, Broadcom, Cisco, Google, HPE, Intel, Meta, Microsoft2024 / v1.0 in 2025
SUE (Scale-Up Ethernet)Broadcom2024
ICIGoogleTPU v4 era (~2018)
SlingshotHPE (Cray)2019

Next: Congestion Control Options → — the algorithms that pair with these transports.