Your Network Is Telling You Something. Are You Listening?

Dec. 1, 2020
Performance Data Analysis in the Era of Working and Learning Online — Millions of Americans are currently working and learning from home. Before working outside of the office was the […]

Performance Data Analysis in the Era of Working and Learning Online —

Millions of Americans are currently working and learning from home. Before working outside of the office was the norm, only around 7% of US workers had the option to regularly work from home. In this new day and age, what is the "norm"? According to a CNBC report from earlier this year, while 55% of Americans will want to work in an office environment at some point at the same or limited capacity, once the economy reopens, 24% say they’d like to work either entirely or more from home compared to how they worked before. That’s a large jump from pre-COVID-19 days.

With over 53% of Americans considering the Internet as "essential" during the pandemic, it’s imperative that providers ensure that reliable Internet access is available at all times. How can providers guarantee workers and students reliable Internet service at home? The answer can be found in the network performance data.

Food for Thought from Our 2022 ICT Visionaries

There are 3 congestion points that are likely to hinder the subscriber’s experience of peak Internet performance:
1. Home Wi-Fi network
2. Outside access network leading to the home
3. Provider’s aggregation network

Application Visibility

In the home, the Wi-Fi network is increasingly the source of data congestion, subscriber frustration, and service provider blame. A poor Wi-Fi network can ruin a brilliantly fast broadband network in the eyes of the subscriber.

During our time of work and study from home we’re using more of our home Internet bandwidth on applications and activities we would have previously conducted at businesses and schools. Video meeting company Zoom now has over 265,000 customers with more than 10 employees each, a number that’s grown 354% since the COVID-19 pandemic began earlier this year. That’s a lot of video meetings that would have been in person at the office or in a school’s classroom.

Gaming has also seen a rise since we began quarantining at home. According to a WeForum study, gaming traffic during peak hours increased 75%. Tools to watch gaming also saw an increase in users. Twitch, YouTube Gaming, and Facebook Gaming are all experiencing growth, with an approximate 20% increase in usage hours reported across services.

And clearly more people are streaming and binging at all hours of the day. The new average is about 8 hours per day — more than double the number of hours before the pandemic. That means bandwidth capacity on home Wi-Fi networks is strained.

While all these activities take bandwidth, applications are not the same in how they use data. Binging a Netflix series will require up to 16 Mbps of near continuous download capacity while high bursts of upload speed will be necessary when collaborating live on a creative student film project or choir practice.

Network performance data within the home Wi-Fi network can help service providers understand how their subscriber’s upload and download speeds are utilized and changing over time. As we continue to work and learn from home, we’re participating in activities that depend as much on upload speeds as download, including taking part in virtual meetings and participating in real-time interactive collaboration.

In an ideal world, an Internet provider should review Wi-Fi performance data and analyze what’s being used and at what hours. With a focus on the application data traversing the home network, a provider could give Zoom meetings, for instance, highest priority, therefore ensuring that Wi-Fi bandwidth is used to support a video conference work meeting or online classroom rather than shared equally with a streaming 4K Netflix binge.

It is critical for providers to expand beyond I deliver Wi-Fi coverage across the entire home and begin looking at the application data necessary to ensure a high subscriber quality of experience (QoE).

Looking Deeper Into Access Network

Congestion caused by the outside access network’s physical limits can likewise result in a poor Internet access experience for the subscriber. No matter how a home is connected to the service provider local office or headend — be it over fiber, HFC, DSL, or fixed wireless — performance data can show if the network can handle the aggregate capacity of all the subscribers.

It’s not as simple to upgrade the outside access network as it is to fine-tune a home Wi-Fi network, but small changes can be made that result in big improvements. For example, a subscriber on a large 2.5G GPON network may be capacity-constrained not by their individual use of the PON bandwidth, but by the aggregate demands of all 64 subscribers on the PON. In this case, the service provider may be able to quickly split the PON into two 32:1 PONs or overlay the GPON network with an XGS-PON network to shift power users to the new higher capacity network.

A VDSL service may be capacity-limited because the copper loop is too long, requiring the service provider to deploy fiber deeper into the network, shortening the copper connection to the home.

Similarly, a fixed wireless network connection may be improved by shorting the distance between base station and the home, or by adding additional wireless spectrum.

Each of these access network remedies are not new, and many have been in progress for years as service providers look to increase capacity to keep up with the incremental year-over-year increase in bandwidth use. The COVID-19 pandemic has put additional strain on the access network, particularly in the upstream direction where DSL, HFC, and wireless networks, are inherently limited. A thorough review of access network performance data shows where and how to address the new outside network bottlenecks.

Starting and Ending at the Top

Most service provider networks can be viewed as a hierarchical distribution and aggregation network — a bit of a pyramid shape as myriad numbers of homes connect to access networks, access networks into aggregation networks, and aggregation networks into regional and core networks.

Continuous monitoring and analysis of network performance data at all levels of the aggregation and content delivery network can identify congestion points that have a wide scope of subscriber impact. Service providers around the globe have been upgrading their aggregation switch, router, and transport capacity, to satisfy the COVID-19-pandemic-fueled hunger for bandwidth.

With insight into network performance data, a provider can understand why the network is performing the way it is, and can ensure that QoE-impacting congestion is limited and quickly removed.

Like this Article?

Subscribe to ISE magazine and start receiving your FREE monthly copy today!

In the era of COVID-19, we are spending more time online with essential work, education, e-commerce, and, yes, even entertainment activities. Not only is congestion more prone to happen but the impact is more damaging.

The analysis of network performance data can help solve old problems, and create new opportunities for subscribers and service providers. It’s time to take a closer look at what the data has to tell us.

References and Notes

About the Author

Alan DiCicco

Alan DiCicco is Solutions Marketing Senior Director, Calix. He is a solutions marketing and technology evangelist responsible for helping service providers deliver new services that provide exceptional ROI. Alan’s 20 years of networking and communications experience includes senior roles in solutions marketing and product management for optical access systems, wireless solutions, and software platforms. He also spent the first few years of his communication career working for the Jet Propulsion Laboratory (NASA), and is passionate about planetary exploration. For more information, please visit For more information about Calix Cloud, please visit