Beyond Bandwidth: Why Latency is the Key to a Seamless Internet Experience
March 29, 2023
Latency, the time it takes for data to travel from one point to another, is often overlooked in the broadband industry. While bandwidth has been the primary focus, latency is just as crucial, if not more so, in delivering an excellent application performance for customers. This was the main message from Gavin Young, Head of Fixed Access Center of Excellence at Vodafone Group, during ourUnderstanding Latencywebinar series, a three-day knowledge sharing event that took place from March 6th to March 8th.
The series brought together world-renowned experts on network latency, including application/OS developers, network vendors, Operators, and ISPs, to share their experiences, insights, and best practices related to measuring and improving latency.
Latency Reduction: An Imperative for Optimal Network Performance and Customer Satisfaction
One of the key takeaways from the industry experts was that reducing latency is critical for creating a network that is invisible to customers and provides the right resources to deliver excellent application outcomes every time they use it. Several speakers and panellists emphasised that insufficient bandwidth results in latency, and thus, latency is not simply one dimension of network quality, it is network quality.
Improving Network Performance through Smarter Queuing and RPM Measurement
Excessive network latency can cause frustration and waste time. Smarter queuing, in which the network manages queues and sends explicit congestion notifications to senders instead of dropping packets, can help alleviate the problem. Fixing the bottleneck queue in the network can make a significant difference in user experience, even if other links have buffer bloat.
Anonher important consideration in measuring network performance is the use of round trips per minute (RPM) as a metric, rather than milliseconds. Stuart Cheshire at Apple emphasised the importance of RPM as a compact integer representation that allows for easy comparison of different network equipment or ISPs.
Educating End-Users on the Benefits of Low Latency
During one of the panel discussions, Gino Dion, Head of Innovation at Nokia, Magnus Olden, CTO at Domos, and Angus Laurie-Pile, Technical Lead at GameBench, highlighted the need to educate end-users on the benefits of low latency. Nokia has found that consumers are willing to pay for a better quality of service experience, but the panellists pointed out that there is still a need for more education on what this quality of service entails and how it can benefit different use cases. The discussion also emphasised the significance of consistent latency in delivering great interactive experiences - it’s not just about the absolute latency, but how stable and reliable the latency is.
The Solution to Improve Network Performance
To tackle the challenges discussed, ISPs should adopt advanced technologies to improve network performance, especially when it comes to reducing latency. L4S, FQ, LLD and LL-APIs are some of the solutions that were highlighted to optimise latency-sensitive traffic.
L4S, or Low Latency, Low Loss, Scalable Throughput, is a protocol designed to reduce latency and packet loss by fixing the bufferbloat problem. FQ, or Fair Queuing, is another technology that assigns a fair share of bandwidth to each user, preventing any one user from monopolising the network. LLD, or Low Latency DOCSIS, is an improvement to the DOCSIS standard for delivering internet over CoAX cables, that improves performance for latency-sensitive traffic such as voice and video. LL-APIs, or Low Latency Application Programming Interfaces, allow applications to communicate directly with the network and make use of technologies to optimise latency-sensitive traffic.
Conclusion: The Crucial Role of Latency in Delivering a Seamless Online Experience
In conclusion, the Understanding Latency webinar series brought together industry leaders such as Gavin Young (Vodafone), Stuart Cheshire (Apple), Dave Taht (Bufferbloat Project), Gino Dion (Nokia) and Magnus Olden (Domos) to underscore the critical role of latency in providing a seamless online experience for end-users. The series emphasised the significance of employing smarter queuing, fair queuing algorithms, and network responsiveness measurement to minimise queuing delay and improve network performance. The need to educate end-users on the advantages of low latency and the opportunity for ISPs to provide low-latency service, was also highlighted. With the internet continuously evolving, reducing latency is becoming increasingly pivotal in delivering excellent outcomes for customer applications and creating a frictionless online experience.
Get the full recordings and learn more at understandinglatency.com.
Latency, the time it takes for data to travel from one point to another, is often overlooked in the broadband industry. While bandwidth has been the primary focus, latency is just as crucial, if not more so, in delivering an excellent application performance for customers. This was the main message from Gavin Young, Head of Fixed Access Center of Excellence at Vodafone Group, during ourUnderstanding Latencywebinar series, a three-day knowledge sharing event that took place from March 6th to March 8th.
The series brought together world-renowned experts on network latency, including application/OS developers, network vendors, Operators, and ISPs, to share their experiences, insights, and best practices related to measuring and improving latency.
Latency Reduction: An Imperative for Optimal Network Performance and Customer Satisfaction
One of the key takeaways from the industry experts was that reducing latency is critical for creating a network that is invisible to customers and provides the right resources to deliver excellent application outcomes every time they use it. Several speakers and panellists emphasised that insufficient bandwidth results in latency, and thus, latency is not simply one dimension of network quality, it is network quality.
Improving Network Performance through Smarter Queuing and RPM Measurement
Excessive network latency can cause frustration and waste time. Smarter queuing, in which the network manages queues and sends explicit congestion notifications to senders instead of dropping packets, can help alleviate the problem. Fixing the bottleneck queue in the network can make a significant difference in user experience, even if other links have buffer bloat.
Anonher important consideration in measuring network performance is the use of round trips per minute (RPM) as a metric, rather than milliseconds. Stuart Cheshire at Apple emphasised the importance of RPM as a compact integer representation that allows for easy comparison of different network equipment or ISPs.
Educating End-Users on the Benefits of Low Latency
During one of the panel discussions, Gino Dion, Head of Innovation at Nokia, Magnus Olden, CTO at Domos, and Angus Laurie-Pile, Technical Lead at GameBench, highlighted the need to educate end-users on the benefits of low latency. Nokia has found that consumers are willing to pay for a better quality of service experience, but the panellists pointed out that there is still a need for more education on what this quality of service entails and how it can benefit different use cases. The discussion also emphasised the significance of consistent latency in delivering great interactive experiences - it’s not just about the absolute latency, but how stable and reliable the latency is.
The Solution to Improve Network Performance
To tackle the challenges discussed, ISPs should adopt advanced technologies to improve network performance, especially when it comes to reducing latency. L4S, FQ, LLD and LL-APIs are some of the solutions that were highlighted to optimise latency-sensitive traffic.
L4S, or Low Latency, Low Loss, Scalable Throughput, is a protocol designed to reduce latency and packet loss by fixing the bufferbloat problem. FQ, or Fair Queuing, is another technology that assigns a fair share of bandwidth to each user, preventing any one user from monopolising the network. LLD, or Low Latency DOCSIS, is an improvement to the DOCSIS standard for delivering internet over CoAX cables, that improves performance for latency-sensitive traffic such as voice and video. LL-APIs, or Low Latency Application Programming Interfaces, allow applications to communicate directly with the network and make use of technologies to optimise latency-sensitive traffic.
Conclusion: The Crucial Role of Latency in Delivering a Seamless Online Experience
In conclusion, the Understanding Latency webinar series brought together industry leaders such as Gavin Young (Vodafone), Stuart Cheshire (Apple), Dave Taht (Bufferbloat Project), Gino Dion (Nokia) and Magnus Olden (Domos) to underscore the critical role of latency in providing a seamless online experience for end-users. The series emphasised the significance of employing smarter queuing, fair queuing algorithms, and network responsiveness measurement to minimise queuing delay and improve network performance. The need to educate end-users on the advantages of low latency and the opportunity for ISPs to provide low-latency service, was also highlighted. With the internet continuously evolving, reducing latency is becoming increasingly pivotal in delivering excellent outcomes for customer applications and creating a frictionless online experience.
Get the full recordings and learn more at understandinglatency.com.
