L4S

L4S (for Low Latency, Low Loss and Scalable Throughput) is an IETF network protocol and congestion control technology designed to simultaneously lower network latency and packet loss rates by reducing bufferbloat throughout the Internet, while preserving network throughput. It uses novel congestion control mechanisms to reduce queuing in the network.[1]

|L4S effectively introduces new rules for compliant endpoints and their traffic, giving L4S traffic preferential treatment in exchange for L4S endpoints cooperating by using improved congestion control algorithems. It has the remarkable property of not only improving performance for L4S traffic, but also improving performance for non-L4S traffic sharing the same infrastructure.

L4S has the advantage that it is an incremental technology which can start to provide incremental latency and throughput improvements through patchwork deployment by individual network operators without having to be adopted throughout the entire Internet.[2]

Details

[edit]

L4S uses Explicit Congestion Notification (ECN) to transmit information about path latency problems, and allows congested nodes to use the ECN bits to send information back to senders that will allow them to adjust their transmit rate, reducing the need for data buffering within router queues.

L4S is specified in RFC 9330. It uses the last codepoint of the Internet Protocol header's ECN field that had not previously been assigned to signal that traffic is from an L4S-capable sender.[3] The full set of four ECN codes for packets are thus:[4]

Code Meaning
00 Non-ECN
01 L4S ECN
10 Classic ECN
11 Marked as congested

Routers can thus treat L4S traffic differently from non-L4S traffic, knowing that L4S endpoints will respond to throttle back traffic in a more controlled way than would be possible using classic ECN. This is done by treating L4S traffic differently for both the cases of queuing and marking.[5]

In an ideal world, L4S traffic would not need any network traffic policing, being entirely self-regulating. In practice. policing may be required to prevent attacks on infrastructure from mis-labeled traffic being introduced by non-compliant endpoints.[6]

Deployment

[edit]

As of January 2025, Internet service providers had started to roll out L4S in their production networks, with Comcast being an early adopter.[7] Apple have incorporated L4S support in their newer operating systems since 2023.[8] Linux support for L4S, in the form of TCP Prague, is available on an experimental basis, and is expected to be merged into the main Linux kernel tree soon.[5]

As of July 21, 2025 T-Mobile announced support for L4S at a network level.[9]

See also

[edit]

References

[edit]
  1. ^ Jackson, Mark (2024-09-19). "Broadband Forum Push Home ISPs to Adopt Low Latency L4S Technology". ISPreview UK. Retrieved 2025-01-29.
  2. ^ Cracknell, Tiffany (2024-09-19). "2024.09.19 - Roadmap launched to implement low latency 'L4S' technology into broadband networks". Broadband Forum. Retrieved 2025-01-29.
  3. ^ Schepper, Koen De; Briscoe, Bob (January 2023). The Explicit Congestion Notification (ECN) Protocol for Low Latency, Low Loss, and Scalable Throughput (L4S) (Report). Internet Engineering Task Force.
  4. ^ "Implementing the 'Prague Requirements for Low Latency Low Loss Scalable Throughput (L4S)" (PDF). bobbriscoe.net. Retrieved 2025-01-31.
  5. ^ a b "Understanding Prague for L4S". Retrieved 2025-01-31.
  6. ^ Briscoe, Bob; Schepper, Koen De; Bagnulo, Marcelo; White, Greg (January 2023). Low Latency, Low Loss, and Scalable Throughput (L4S) Internet Service: Architecture (Report). Internet Engineering Task Force.
  7. ^ Peters, Jay (2025-01-29). "Comcast is rolling out 'ultra-low lag' tech that could fix the internet". The Verge. Retrieved 2025-01-29.
  8. ^ Clark, Mitchell (2023-12-09). "The quiet plan to make the internet feel faster". The Verge. Retrieved 2025-01-29.
  9. ^ Saw, John (July 21, 2025). "T‑Mobile Is First to Unlock L4S in Wireless — A Key Step Toward a Smarter, Programmable 5G". T-Mobile.
[edit]