Low Latency, Low Loss, Scalable Throughput (L4S): A Technical Overview

L4S is about to make the internet feel faster than anything that’s come before it.

Written by

Last Updated: Jan 2, 2024
Abstract image representing L4S technology.
L4S could transform the way the web “feels” in the coming years.
  • L4S technology marks a substantial improvement in internet speed and reliability, critical for applications requiring real-time interaction, such as online gaming and video conferencing.
  • By intelligently handling how data packets are processed and reducing network congestion, L4S minimizes delays often caused by bufferbloat, leading to smoother and more efficient data transmission, particularly noticeable during high-traffic times.
  • L4S is not just a solution for current network challenges, but also a framework designed for future technological advancements. L4S’ scalability ensures networks are ready to handle increasing and evolving data requirements.

L4S, or Low Latency, Low Loss, Scalable Throughput, is a transformative network technology standard aimed at enhancing internet traffic performance. Finalized in January 2023, L4S addresses the growing need for efficient, real-time data transmission essential for applications like online gaming, video conferencing, and VoIP. The technology has garnered significant interest from major companies, including Apple, Google, Comcast, and Nvidia​​.

Why L4S Matters

The advent of L4S technology marks a significant milestone in the evolution of internet connectivity, addressing critical challenges faced by modern digital communications. Here’s why it’s important:

  • Enhanced real-time communication: L4S drastically reduces latency, a crucial factor for applications like online gaming, video conferencing, and telemedicine. This enhancement leads to smoother, more reliable real-time interactions.
  • Improved network efficiency: By implementing scalable throughput and advanced AQM (active queue management) techniques, L4S optimizes data flow, reducing congestion and packet loss even during high-traffic periods.
  • Future-ready infrastructure: As the Internet of Things (IoT) and 5G technologies grow, L4S provides a scalable and efficient framework, ensuring that network infrastructure can handle increasing data demands.
  • User experience: L4S directly contributes to a better user experience by providing faster, more consistent internet performance.

Historical Development of L4S

L4S technology began as an initiative to address the limitations of traditional TCP/IP networking, particularly in handling high-bandwidth and low-latency demands. This evolution was marked by significant research into AQM and congestion control mechanisms.

Early efforts focused on improving the responsiveness and stability of internet traffic, eventually leading to the formulation of the L4S standard. Milestones in this journey include the development of foundational AQM techniques and the integration of scalable congestion control protocols, culminating in the standardization of L4S in January 2023.

Deployment and Compatibility of L4S

L4S is designed for incremental deployment alongside existing best-effort services. This approach facilitates easier integration into current network infrastructures. However, deploying L4S requires certain network features, particularly AQM systems compatible with its requirements. Therefore, existing network hardware, including routers, may need upgrades or specific configurations to support these systems. Newer routers or those receiving firmware updates from manufacturers are more likely to be compatible with L4S standards​​​​​​.

Many major brands and organizations are already signed on to bring L4S support, including Apple, which notably added support for the feature starting with iOS 16 and MacOS Ventura at WWDC 2023.

Technical Challenges and Solutions for L4S

L4S faced numerous technical challenges, notably integrating with existing network infrastructures and ensuring compatibility with a wide range of applications. A major hurdle was designing an AQM system capable of differentiating between L4S and classic traffic efficiently.

Solutions involved the development of the Dual Queue Coupled AQM system and advancements in scalable congestion control algorithms. These innovations helped L4S achieve its goal of reducing latency and packet loss without compromising throughput.

Bufferbloat and How It Relates to L4S

Technical diagram showcasing one of the benefits of L4S
L4S helps minimize latency by handling packets in a more dynamic way, tracking how long they are held onto and therefore reducing instances of bufferbloat. (Image: Nokia Bell Labs)

Bufferbloat is a critical issue in network communication, characterized by excessive latency and jitter due to overfilled buffers in routers and switches. This phenomenon occurs when network devices hold onto data packets for too long, causing delays and a decrease in overall network performance, particularly impacting activities requiring real-time data transmission like video conferencing and online gaming.

The primary cause of Bufferbloat is the mismatch between the capacity of network paths and the size of buffers in networking equipment. Modern networks often deploy buffers larger than necessary, operating on the assumption that a “bigger buffer” equates to better performance. However, this results in extended data queuing times, leading to increased latency and jitter.

L4S technology directly addresses bufferbloat by implementing intelligent queue management. L4S introduces a more dynamic approach to handling network traffic, significantly reducing latency and improving overall network efficiency. It uses advanced AQM techniques, which dynamically adjust buffer usage based on current network conditions, ensuring that buffers do not become overfilled and thus mitigating the effects of bufferbloat.

Furthermore, L4S differentiates between various types of traffic, prioritizing latency-sensitive applications. This ensures that essential data packets are not delayed by less critical traffic, further alleviating the impact of bufferbloat.

Despite the effectiveness of L4S in combating bufferbloat, challenges remain in its widespread implementation. Network infrastructure upgrades and the adoption of L4S-compatible protocols are necessary for its full potential to be realized. Nonetheless, the ongoing development and integration of L4S technology represent a significant step toward more efficient and reliable network communication, offering a promising solution to the long-standing issue of bufferbloat.

L4S Technical Architecture

L4S operates on three main components:

  • Scalable congestion control on the sending host: Scalable congestion controls, like Data Center TCP (DCTCP), have been implemented in various environments, including Windows Server Editions, Linux, and FreeBSD. These controls address the scaling issue with classic congestion controls (like Reno or CUBIC), ensuring more consistent and efficient data flow​​​​.
  • AQM at the network bottleneck: Dual Queue Coupled AQM is employed to partition latency between L4S and classic traffic without dividing bandwidth. This system enables the network to manage traffic more efficiently, reducing latency and packet loss​​​​.
  • Protocol features for traffic identification: The Explicit Congestion Notification (ECN) protocol is used for communication between network elements and hosts. This allows for identifying L4S traffic, ensuring it receives the intended low-latency, low-loss service​​​​.

What Does L4S Mean for the Future?

The potential future developments of L4S are vast, with implications for various sectors. Researchers are exploring enhancements that could further reduce latency and increase scalability. Integration with emerging technologies like the IoT and 5G networks is a key area of interest. Future iterations of L4S might also focus on increased security measures and energy efficiency, aligning with the growing emphasis on sustainable technology practices.

The L4S standard represents a major advancement in network technology. Its approach to managing internet traffic aligns with the growing demand for high-performance, real-time data transmission. As the technology becomes more widely adopted and supported by network hardware, its impact on various online applications and services is expected to be substantial, marking a significant step forward in the evolution of internet technology.

Frequently Asked Questions About L4S

Is L4S a replacement for current TCP/IP protocols?

No. L4S is not a direct replacement, but rather an enhancement that works alongside existing protocols to improve performance.

Does implementing L4S require significant changes to existing network infrastructure?

While some changes are necessary, particularly in supporting advanced AQM systems, L4S is designed for incremental deployment, allowing for gradual integration with existing networks.

Can L4S be used for all types of internet applications?

L4S is particularly beneficial for applications requiring low latency and high reliability, such as online gaming and video conferencing. However, its benefits can be extended to a wide range of internet applications.

Is L4S technology only for large-scale or enterprise networks?

L4S is scalable and can benefit large-scale and smaller setups, including home networks.