HFC Network

Hourly Data Consumption of Popular Video Conferencing Applications

Doug Jones
Principal Architect

May 6, 2021

Building on our prior work, this investigation explores the hourly data consumption of popular video conferencing applications: Google Meet, GoToMeeting, Microsoft Teams and Zoom. As video conference applications have become an integral part of our daily lives, we wanted to not only better understand the bandwidth usage as previously explored, but also the total data consumption of these applications. This investigation provides a first step in better understanding that latter dimension. To avoid any appearance of endorsement of a particular conferencing application, we have not labeled the figures below with the specific apps under test. In short, we observed that a single user on a video conferencing application consumed roughly one gigabyte per hour, which compares to about three gigabytes per hour when streaming an HD movie or other video. However, we did observe substantial variance in video conferencing app hourly data consumption based on the specific app and end-user device.

Key Components of the Testing Environment

Much like our prior work on bandwidth usage, the test setup used typical settings and looked at both upstream and downstream data consumption from laptops connected to a cable broadband internet service. We used the same network equipment from November and our more recent blog post in February. This includes the same cable equipment as the previous blogs — the same DOCSIS 3.0 Technicolor TC8305c gateway, supporting eight downstream channels and four upstream channels, and the same CommScope E6000 cable modem termination system (CMTS). The cable network was configured to provide 50 Mbps downstream and five Mbps upstream broadband service, overprovisioned by 25 percent.

The data gathering scenario:

  1. 10 people, each on their individual laptops, participated in the conference under test
  2. One person on the broadband connection under test, using either a lower-cost or a higher-cost laptop. The other nine participants were not using the broadband connection under test.
  3. For the laptop under test, the participant used the video conferencing application for the laptop’s operating system, rather than using the video conferencing application through the web browser.
  4. Total data consumption was recorded for the laptop using the broadband connection under test.

For all 10 participants, cameras and microphones were on. Conference applications were set to "gallery mode" with thumbnails of each person filling the screen, no slides were presented and the video conference sessions just included people talking.

The laptop under test used a wired connection to the cable modem to ensure that no variables outside the control of the service provider would impact broadband performance. Most notably, by using a wired connection, we removed the variable of Wi-Fi performance from our test setup. During data collection, the conference app was the only app open on the laptop under test.

Video conferencing sessions were set up and data consumption was measured over time. We collected 10 minutes of data for each conferencing session under test to calculate the total consumption for one hour. The charts below show the data consumed for each of the 10 minutes of the conference session. During the conference there was movement and discussion to keep the video and audio streams active throughout the period of data collection.

For each test scenario, only one laptop was connected at a time to the broadband connection under test. Our goal was to measure the data consumption of one conferencing user on the broadband connection. The other conference participants were on the internet; they were not in the lab. Once again, we used TShark (a popular, widely used network protocol analyzer) to capture and measure the data.

For the laptop under test, we chose two that have quite different capabilities. The first was a low-cost laptop with an 11-inch screen, like the ones students are often provided by school districts for at-home learning. The second was a higher-cost laptop with a 15-inch screen, like what we often see in an enterprise environment. Note the two laptops not only have quite different hardware components (e.g., CPU, graphics processors, memory, cameras, screens), but also have different operating systems. Once again, to avoid any appearance of endorsement, we are not identifying the specific laptops used.

Analysis

Table 1 shows hourly bandwidth consumption (combining both upstream and downstream) for the laptop under test, normalized to Gigabytes per hour. The table provides the data consumption for the low-cost and higher-cost laptops in each scenario with the four conferencing applications.

Table 1: Video Conferencing App Hourly Bandwidth Consumption in Gigabytes for Each User (Gigabytes/hour)

Table 1: Video Conferencing App Hourly Bandwidth Consumption in Gigabytes for Each User (Gigabytes/hour)
 

The following figures show the data consumption, in Megabytes, for each minute of the 10-minute data collection for each of the permutations of our testing.

A few notes on the charts:

  • There was only one client behind the cable modem.
  • Each bar represents one minute of data consumption.
  • Each bar shows total consumption and includes both the upstream and downstream, and both audio and video, added together.
  • App A is blue in each chart; App B is green; App C orange; and App D is purple.
  • These charts show real-time consumption measured in Megabytes per hour to illustrate consumption over time.

Figure 1 shows the data consumed when using the lower-cost laptop in the 10-person meetings.

Figure1

Figure 2 shows data consumed each minute for each of the four apps when using the higher-cost laptop was in the 10-person meetings.

Figure2

Figure 3 shows the data consumed each minute using App A and compares the two laptops used for data collection. For each minute, the bar to the left is the lower-cost laptop and the bar to the right is the higher-cost laptop.Figure3

Figure 4 shows the data consumed each minute using App B and compares the two laptops. The bar to the left is the lower-cost laptop and the bar to the right is the higher-cost laptop.

Figure4

Figure 5 shows the data consumed each minute using App C and compares the two laptops. The bar to the left is the lower-cost laptop and the bar to the right is the higher-cost laptop.

Figure5

Figure 6 shows the data consumed each minute using App D and compares the two laptops. The bar to the left is the lower-cost laptop and the bar to the right is the higher-cost laptop.

Figure6

Key Observations

A. Data Consumption Varies: The first takeaway is that different apps consume different amounts of bandwidth, as shown in Table 1, from 0.5 GBytes per hour up to 3.4 GBytes per hour, for video conferences using the different laptops, the same broadband connections, the same general setup (e.g., gallery view), the same people doing the same things on camera, etc.

    1. For a  given app on a given laptop, data consumption was consistent over the 10-minute collection time.
    2. App D using the higher-cost laptop consumed the most bandwidth.
    3. With App D on the lower-cost laptop, there was video quality degradation. We confirmed the broadband connection was operating as expected and was not the cause of the video degradation. Rather, it appeared that the combination of the hardware and operating system of the lower-cost laptop was unable to meet the resource requirements of App D.
    4. App B consistently consumed less bandwidth regardless of scenario.

B. Comparing Laptops: In Table 1, the two columns of data show the differences between the lower-cost and higher-cost laptops for the data collections. On the lower-cost laptop, Apps A, B and C consume about the same amount of data on an hourly basis.

C. Comparing Laptops: The second column of data show that all apps on the higher-cost laptop consumed more bandwidth than the lower-cost laptop. This difference implies that when using the actual conferencing app (not a web browser), processing power available in the laptop may be a determining factor in consumption.

D. Comparing Apps: App C was the most consistent in data consumption regardless of the laptop used. The other conference applications noticeably consumed more on the higher-cost laptop.

In summary, we observed a more than 7X variation in the data consumption of video conferencing with a very limited exploration of just two variables – laptop and video conferencing application. Notably, however, when data consumption was at its highest, it was of the same magnitude as the data consumption of an HD video stream.

This is an area ripe for further research and study, both to more comprehensively explore these variables (e.g., other device types, larger meetings) and to explore other variables that may meaningfully influence data consumption.

LEARN MORE

HFC Network

Expanded Testing of Video Conferencing Bandwidth Usage Over 50/5 Mbps Broadband Service

Jay Zhu
Senior Engineer

Feb 19, 2021

As working from home and remote schooling remain the norm for most of us, we wanted to build on and extend our prior investigation of the bandwidth usage of popular video conferencing applications. In this post, we examine the use of video conferencing applications over a broadband service of 50 Mbps downstream and 5 Mbps upstream (“50/5 broadband service”). The goal remains the same, looking at how many simultaneous conferencing sessions can be supported on the access network using popular video conferencing applications. As before, we examined Google Meet, GoToMeeting, and Zoom, and this time we added Microsoft Teams and an examination of a mix of these applications. To avoid any appearance of endorsement of a particular conferencing application, we haven’t labeled the figures below with the specific apps under test.

We used the same network equipment from November. This includes the same cable equipment as the previous blog -- the same DOCSIS 3.0 Technicolor TC8305c gateway, supporting 8 downstream channels and 4 upstream channels, and the same CommScope E6000 cable modem termination system (CMTS).

The same laptops were also used, though this time we increased it to 10 laptops. Various laptops were used, running Windows, MacOS and Ubuntu – nothing special, just laptops that were around the lab and available for use. All used wired Ethernet connections through a switch to the modem to ensure no variables outside the control of the broadband provider would impact the speeds delivered (e.g., placement of the Wi-Fi access point, as noted below). Conference sessions were set up and parameters varied while traffic flow rates were collected over time.  Throughout testing, we ensured there was active movement in view of each laptop’s camera to more fully simulate real-world use cases.

As in the previous blog, this research doesn’t take into account the potential external factors that can affect Internet performance in a real home -- from the use of Wi-Fi, to building materials, to Wi-Fi interference, to the age and condition of the user’s connected devices -- but it does provide a helpful illustration of the baseline capabilities of a 50/5 broadband service.

As before, the broadband speeds were over-provisioned. For this testing, the 50/5 broadband service was over provisioned by 25%, a typical configuration for this service tier.

First things first: We repeated the work from November using the 25/3 broadband service. And happily, those results were re-confirmed. We felt the baseline was important to verify the setup.

Next, we moved to the 50/5 broadband service and got to work. At a high level, we found that all four conferencing solutions could support at least 10 concurrent sessions on 10 separate laptops connected to the same cable modem with the aforementioned 50/5 broadband service and with all sessions in gallery view. The quality of all 10 sessions was good and consistent throughout, with no jitter, choppiness, artifacts or other defects noticed during the sessions. Not surprisingly, with the increase in the nominal upstream speed from 3 Mbps to 5 Mbps, we were able to increase the number of concurrent sessions from the 5 we listed in the November blog to 10 sessions with the 50/5 broadband service under test.

The data presented below represents samples that were collected every 200 milliseconds over a 5-minute interval (300 seconds) using tshark (the Wireshark network analyzer).

Conferencing Application: A

The chart below (Figure 1) shows total access network usage for the 10 concurrent sessions over 300 seconds (5 minutes) while using one of the above conferencing applications. The blue line is the total downstream usage, and the orange line is total upstream usage. Note that the total upstream usage stays around 2.5 Mbps which may be a result of running 10 concurrent sessions. Also, the downstream usage stays, on average, around 15 mbps, which leaves roughly 35 Mbps of downstream headroom for other services such as streaming video that can also use the broadband connection at the same time.

Figure 1 - App A total
 

Figure 2 shows the upstream bandwidth usage of the 10 concurrent sessions and it appears that these individual sessions are competing amongst themselves for upstream bandwidth. However, all upstream sessions typically stay well below 0.5 Mbps -- these streams are all independent, with the amount of upstream bandwidth usage fluctuating over time.

Figure 2 - App A up
 

Figure 3 shows the downstream bandwidth usage for the 10 individual conference sessions. Each conference session typically uses between 1 to 2 Mbps. As previously observed with this application, there are short periods of time when some of the sessions use more downstream bandwidth than the typical 1 to 2 Mbps.

Figure 3 - App A down

Conferencing Application: B

Figure 4 shows access network usage for 10 concurrent sessions over 300 seconds (5 minutes) for the second conferencing application tested. The blue line is the total downstream usage, and the orange line is total upstream usage. Note that the total upstream usage hovers around 3.5 Mbps.  The total downstream usage is very tight, right above 10 Mbps.

Figure 4 - App B total
 

Figure 5 shows the upstream bandwidth usage of the 10 individual conference sessions where all but one session is well below 1 Mbps and that one session is right at 2 Mbps.  We don’t have an explanation for why that blue session is so much higher than the others, but it falls well within the available upstream bandwidth.

Figure 5 - App B up
 

Figure 6 shows the downstream bandwidth usage for the 10 individual conference sessions clusters consistently around 1 Mbps.

Figure 6 - App B down

Conferencing Application: C

Figure 7 shows access network usage for the 10 concurrent sessions over 300 seconds (5 minutes) for the third application tested. The blue line is the total downstream usage, and the orange line is total upstream usage. Note that the total upstream usage hovers right at 3 Mbps over the 5 minutes.

Figure 7 - App C total
 

Figure 8 shows the upstream bandwidth usage of the 10 individual conference sessions where all stay well below 1 Mbps.

Figure 8 - App C up
 

Figure 9 shows the downstream bandwidth usage for the 10 individual conference sessions. These sessions appear to track each other very closely around 2 Mbps, which matches Figure 7 showing aggregate downstream usage around 20 Mbps.

Figure 9 - App C down

Conference Application: D

Figure 10 shows access network usage for the 10 concurrent sessions over 300 seconds (5 minutes) for the fourth application tested. The blue line is the total downstream usage, and the orange line is total upstream usage. Note that the total upstream usage hovers right at 5 Mbps over the 5 minutes, and there is no visible degradation to the conferencing sessions was observed.

Figure 10 - App D total
 

Figure 11 shows the upstream bandwidth usage of the 10 individual conference sessions, where there is some variability in bandwidth consumed per session.  One session (red) consistently uses more upstream bandwidth than the other sessions but remained well below the available upstream bandwidth.

Figure 11 - App D up
 

Figure 12 shows the downstream bandwidth usage for the 10 individual conference sessions. These sessions show two groups, with one group using less than 1 Mbps of bandwidth and the second group using consistently between 2 Mbps and 4 Mbps of bandwidth.

Figure 12 - App D down
 

Running All Four Conference Applications Simultaneously

In this section, we examine the bandwidth usage of all four conferencing applications running simultaneously. The test consists of three concurrent sessions from two of the applications and two concurrent sessions from the other two applications (once again a total of 10 conference sessions running simultaneously). The goal is to observe how the applications may interact in the scenario where members of the same household are using different conference applications at the same time.

Figure 13 shows access network usage for these 10 concurrent sessions over 300 seconds (5 minutes). The blue line is the total downstream usage, and the orange line is total upstream usage. Note that the total upstream usage once again hovers around 5 Mbps without any visible degradation to the conferencing sessions, and the downstream usage is pretty tight right above 10 Mbps.

Figure 13 - all 4 total
 

Figure 14 shows the upstream bandwidth usage of the 10 individual conference sessions where several distinct groupings of sessions are visible. There were 4 different apps running concurrently. One session (red) consumes the most upstream bandwidth at averaging around 2 Mbps, whereas the other sessions use less, and some much less.

Figure 14 - all 4 up
 

Figure 15 shows the downstream bandwidth usage for the 10 individual conference sessions across the four apps and, again, there are different clusters of sessions. Each of the four apps are following their own algorithms.

Figure 15 - all 4 down
 

In summary, with a 50/5 broadband service, each of the video-conferencing applications supported at least 10 concurrent sessions, both when using a single conferencing application and when using a mix of these four applications. In all cases, the quality of the 10 concurrent sessions was good and consistent throughout. The 5 Mbps of nominal upstream bandwidth was sufficient to support the conferencing sessions without visible degradation, and there was more than sufficient available downstream bandwidth to run other common applications, such as video streaming and web browsing, concurrently with the 10 conferencing sessions.

CABLE BROADBAND NETWORK PERFORMANCE

Home Network

Screen Yourself: When It’s Wise to Nix Video in Teleconferences

CableLabs
CableLabs Admin

May 22, 2020

To be seen or not to be seen – that is the question for many of us as we spend our days in online teleconference meetings while working from home.

True, seeing video of your fellow participants can add energy to a meeting and make you feel less isolated. But there are some good reasons – and some embarrassing ones – for toggling off your video feed.

My home broadband isn’t all that broad. Even in the era of plentiful home broadband service, video teleconferencing calls can be a chore if your home connection is running slow. Maybe you moved into a house with a great view of the mountains but lousy Internet access. Or every member of your household is online right now, and your share of the broadband pipe is decidedly narrow. Or it could be that you are required to user a VPN connection, which provides better security but also slows down the video stream and increases latency. Whatever the case, participants’ video feeds jerk and stutter, or worse, all of a sudden you find yourself kicked out of the call queue. Turning off your video feed – and asking others to do so as well – may lighten the bandwidth load so you can continue the meeting.

My laptop is a fossil. Hey, we’ve all been there. You start a job and you are given the oldest laptop in the company fleet. In computer years, it dates back to the Cretaceous Era, as evidenced by its CD drive. With its maxed-out RAM memory and outdated operating system, this geezer of a computer can barely deal with email, let alone a live streaming video conferencing call. If your laptop would be more useful as a brick and there is no hope of a replacement, you may need to participate in teleconference meetings in audio-only mode.

Focus, people! If the meeting features a single person giving a presentation, it may be a good idea to have everyone else go sans video. Not only does this allow the group to focus on the presenter as well as any screen-shared information, but the presenter doesn’t have to watch a half-dozen or more bobbing heads, or see that most of the participants are looking elsewhere, munching on snacks or yawning. That’s not only discouraging for the presenter, but it also is distracting.

Oh no, I really have to go… You are in the middle of a marathon meeting and you really, really need to use the restroom. Or maybe your five-year-old has just burst into your home office in full-on meltdown mode. Either way, it’s probably best if you temporarily turn off your video and put yourself on mute. Your colleagues may not be happy you are taking a break to deal with personal business, but they will be grateful you didn’t share it in glorious video and audio.

Uh, did I put on pants? You overslept. Or you are naturally fashion-challenged. Or the call came at a really bad time early or late in the day. Whatever the case, you aren’t exactly dressed for success – in fact, you shouldn’t even go out in public and risk scaring small children. The best option may be to turn off your computer camera and keep your fashion fail private. All in all, your colleagues may thank you for it. You really don’t need to be this guy.

Subscribe to our blog to get more tips, tricks and information from CableLabs.

SUBSCRIBE TO OUR BLOG