Swimming upstream; Managing the changing trends in network usage during COVID-19

April 16, 2020

As the world works to contain COVID-19, “staying home” has become a reality, and as a result, many more of us have had to adapt to working from home, proving just how important connectivity is during these times. Video helps us to stay connected to colleagues in order to continue working as well as allowing us to communicate with isolated family members. Video has also proved to be an essential tool for e-learning when childrencan’t attend school and for the use of streaming services to keep us entertained in the evenings. These reasons combined mean that operators have notices an upward surge in both upstream and downstream rates.

Upstream data trends, which refers to data originating from the user, are significant as they represent applications such as video conferencing and video calls. These are both essential for keeping businesses moving and keeping us in touch with loved ones.

Drilling down – What are the operators seeing?

According to CommScope, different patterns emerged over the first three weeks of the US lockdown. Week one saw a new bandwidth peak from 11:30 am -12pm. This equaled 75% of their usual evening peak. Week two saw this increase to nearly 100% of the evening peak and stretched from 12 pm – 9 pm, a massive shift. Week three saw these patterns continue as more and more people started working from home and using video conferencing services.

Vodafone noticed huge increases with their most extreme trend coming from upstream data, which they attributed to video conferencing has increased by 100%. Added to this is the use of streaming traffic, which has increased by 50% on fixed broadband across their European networks. These increases have put them at near capacity for their evening peaks, and as a result, they have planned upgrades where they will add four terabits per second of additional capacity to help ease some of the strain.

Virgin Media in the UK also cited a significant jump in usage. They stated their downstream broadband traffic was up 90% on the first day of the school closures. Their upstream traffic has more than doubled during daytime hours, which they too attribute to video calls and conferencing applications. Jeanie York, Virgin’s Chief Technology and Information Officer said, the increase in upstream traffic is linked to more people working from home and the rising trend of video conferencing during social distancing. She continued by clarifying, “This traffic is increasing throughout the day and continuing into the evening, with peak upstream traffic up around 25% on the previous week, showing people are working later or joining conference calls with friends and family.”

As people get used to this new way of living operators will continue to see new trends in times the peaks occur and what applications are seeing the heaviest usage. For example, one such trend that has emerged is upstream traffic, which tends to dip slightly at lunchtime as most people stop to take a lunch break, trying to build structure into their day and enforcing breaks.

Assuring video services

Operators are suggesting that the increase in upstream traffic can be attributed to video traffic. If that is the case and video is one of the main ways people are maintaining communication, both for work and socially, then ensuring the quality of experience is of the highest importance.

Video traffic, which accounts for the highest data usage is also, almost 100% encrypted traffic, which leaves operators in an awkward position if they want to assure the service quality. Therefore, operators must employ a heuristic approach coupled with advanced techniques such as AI and machine learning to try and understand the trends in the video traffic. Once these patterns are understood, the operator can set Key Quality Indicators (KQIs) and measure the Quality of Experience QoE.

However, in order to deliver this level of service, operators need to be employing an assurance solution that allows them to track and optimize video traffic. To collect the necessary data required to understand the patterns, service assurance providers can gather statistics on specific metrics alongside a crawler, which calculates KQI’s and, at the end of the session, will prompt the user to enter the real customer experience observed. Thousands of these samples are required to generate enough information so that machine learning can develop algorithms for identifying and improving these KQI’s. Together these can help measure performance, and if an anomaly occurs or service levels drop below a certain threshold, automatic alarms will be triggered, and operators can troubleshoot on-demand.

Decoding encrypted traffic is, therefore, vital to understanding upstream traffic, including video. If operators want to be delivering the highest possible levels of service and connectivity, then they need to be able to crack the code of encrypted traffic. Using advanced techniques in AI and machine learning, operators can monitor the upstream video KPIs smartly.


As the situation changes daily, and requirements on the network shift, operators need a smart containerized solution which can react in real-time to optimize the network. If degradation occurs for video services, operators must be able to troubleshoot on the fly and drill down for root-cause analysis. If an operator wants to manage the upstream traffic, instead of fighting against the current, they will need a fully virtualized smart, containerized service assurance solution to manage it successfully.

RADCOM Service Assurance is a dynamic, fully containerized, and on-demand solution that uses advanced techniques in AI and machine learning to gain insights into encrypted traffic. To learn more about RADCOMizing your network click here.

Share this article
Skip to content