Video Conferencing Network

Best Practices

Not All Video Is Created Equal

It’s no secret that video conferencing is a productivity powerhouse around the
office. For users, it helps improve employee engagement and creates a better
sense of community among globally dispersed teams. But the IT departments
implementing the solution will find some unique benefits of their own. An all-in-
one solution like Lifesize can replace legacy audio-only conferencing services
and provide audio, video and web conferencing in a single application that tightly
integrates with conference room cameras and touchscreen conference phones.
It’s a simple solution for users to use and for IT to manage. Just remember that
not all video is created equal.

There are two main differences between the video clips you watch
on the internet and the live video conferences you hold at the office.
For one, streaming video clips (your standard cat video variety) are
designed to download portions of the footage ahead of time to make
up for network instability during the course of the playback. The other
is that you almost never raise an IT helpdesk ticket because your
YouTube video took a few extra seconds to buffer. Video conferencing,
on the other hand, relies on packets of data being sent and received
consistently and in real time to ensure that the call remains clear, with
every inflection, gesture and word correctly transmitted.

When researching a video conferencing solution, it’s important to

review your own network topology to see if any optimizations could
be made to ensure the best possible experience for your end users. It
should be noted that most deployments are ready for and run video
perfectly well without any modifications. The tips in this guide are
designed to help you get the most out of your network by outlining
some key considerations for your bandwidth, network architecture and
network security.
 Video Conferencing Network Best Practices

How Much Bandwidth Do I Need?

The first question most IT admins have about video conferencing isn’t
about 1080p resolution or if it supports the latest emoji packs; it’s all about
bandwidth requirements. When we break it down, a point-to-point video call
between two users or locations will use about 2 Mbps down and 2 Mbps up
of internet bandwidth on both sides — since both groups are sending and
receiving packets of data during a live call. These requirements are designed
to make sure the call runs smoothly and with the best quality possible (or
from IT’s perspective, to minimize the chance of a quality-related helpdesk
ticket being submitted).

But unless you deploy a single video conferencing room with a strict “one at a time” rule, you’re
likely to run into situations with concurrent calling. By the way, this is a very good thing, both
for adoption and productivity. Aside from maybe security alarm systems, high utilization is
generally good. According to our data about overall Lifesize service usage, a typical organization
should account for at least 5 percent concurrent utilization, or five simultaneous calls for every
100 users sharing an internet connection. Organizations that think they’ll use the service
significantly more can factor that in. It’s also important to build in an additional 10 percent of
overhead to account for call signaling and packetization.

Video Bandwidth (Mbps) =

( Accounts 2 Mbps .05 ) + 10%

(Users + Conference Room Systems) (Max Call Rate) (Concurrent (Protocol Overhead)
Utilization %)

Now that you have a baseline requirements number, it’s time to look at your current bandwidth
allotment and current load. Unused bandwidth can be an expensive luxury, so most companies
prefer to purchase the exact amount they need. If your video bandwidth requirement fits inside
your unused allocation, you’re good to go. If not, it may be worth a call with your ISP to upgrade
your internet pipe for the best quality.

Video Network Best Practices | 2

Video Conferencing Network Best Practices

Handling Internet Imperfections

Like most things, the public internet has its imperfections. Even if your office
is perfectly equipped with the recommended amount of bandwidth, there
are still times when packets can be dropped due to internet failings — luckily,
we’ve planned ahead. Lifesize incorporates three levels of packet loss and
internet imperfection handling to keep your conversations running smoothly.

Error Concealment
Lifesize runs the very efficient Opus audio codec with built-in error concealment, offering no
noticeable artifacting at up to 20 percent packet loss. Video transmissions are also built with
error concealment to seamlessly rebuild dropped data using the last known frames and pieces
(called micro blocks). Our error concealment technology is the first line of defense and has the
ability to mask dropped packets with almost no visible or audible change.

Error Correction
In the event of a large data loss, error correction will kick in to mathematically rebuild and fill
in gaps based on the most recent frames of data. In these instances, audio is prioritized to
maintain the flow of the conversation while the data is recovered.

Rate Control
Lifesize is built with the ability to dynamically downspeed a call to adjust for bandwidth
limitations. If you are not able to send or receive the recommended 2 Mbps, the call will
dynamically adjust your resolution to make the most of the available bandwidth. This will not
affect other participants on the call who are able to send and receive 2 Mbps; they will continue
to experience the call at full quality. Like error correction, rate control also prioritizes audio in
Real-Time System Health Monitoring the event of internet instability.
Lifesize Icon Health Monitoring utilizes data
analytics to proactively alert your IT team The Lifesize cloud architecture is built to rely as little on the public internet and as little on
to potential errors detected in the network. your bandwidth as possible. When you have Lifesize calls or meetings with participants located
Upon enablement, you’ll get alerts for around the world, instead of traversing the entire distance on the public internet, the call is
anything from disconnected cables to PBX routed to the nearest IBM SoftLayer PoP (point of presence), where it jumps on their ultrafast,
registration errors and instances of excessive managed fiber network to navigate the globe. The call isn’t constantly hopping among different
packet loss. Learn more. providers like it would on the public internet, which would increase latency and produce a
much lower-quality call. The Lifesize and IBM Cloud relationship is designed to take the daily
maintenance and connectivity reliability off of your IT team and give you the best possible video
quality.

Video Network Best Practices | 3

Video Conferencing Network Best Practices

Video Network Architectures

Like bandwidth, your network topology can have a Medium, Large and Enterprise Networks
Larger organizations typically have some kind of WAN that spans
positive, or negative, impact on your overall video
multiple offices or even continents. They may deploy a mixture
experience. Below, we’ve outlined the main types of of IP-VPN and MPLS connectivity interconnecting these sites but
network architectures that we regularly encounter. generally do not have local internet breakouts. Instead, these networks
Depending on factors like the industry and size of often deploy centralized or regional breakouts colocated with other
infrastructure and storage solutions.
your company, there may be recommendations or
requirements for improving video traffic through your This means that users who make calls from offices without an internet
network. breakout will find their traffic routed across the corporate WAN
links, which are often dimensioned to provide minimum acceptable
Small and Medium Business Networks performance due to high cost. This can have a negative impact on
Small companies with one or two office locations typically have call experience if the WAN links are not suitably provisioned with
simple IT environments. They may run independent networks with no enough bandwidth to support your video traffic. Adding Quality of
interconnections, or they may deploy IP-VPN technologies like Internet Service (QoS) to the networks can help prioritize the traffic traversing
Protocol Security (IPsec) VPN tunnels to interconnect their offices and these links, but this only really helps when links are already becoming
users via VPN clients. These networks usually have localized breakouts congested or so heavily utilized that packet loss is occurring.
to the internet at each office location.
Offloading bandwidth-hungry and latency-intolerant real-time traffic at
the local or regional office is less demanding on WAN links. And since
it’s often more cost-effective to expand internet bandwidth than WAN
connectivity, using “the cloud” to carry your real-time communication
traffic is a logical approach and might be worth considering for the
future, if not today.

IPsec
1

Location A Location B
3

This network architecture lends itself well to the use of cloud-based

Location D
solutions, as instead of having to traverse a WAN (wide area Location C
network) connection to reach the internet, the path is much
shorter to the edge of the network. This is particularly
important for real-time applications like video as fewer Location B
hops often mean less latency.

IPsec effectively wraps video data packets in additional security data

1 packets, significantly bloating the size of the call and reducing the
MPLS 2

quality.
Corporate HQ

MPLS pipes are usually significantly smaller than the public internet Your MPLS and IPsec solutions may be fine for connecting network
2
pipe, basically putting a bottleneck on the network, which could drives, but at the end of the day, if it takes a couple extra minutes to
ultimately risk the quality of your video calls. send a file, it’s not a big deal. By contrast, the real-time nature of video
communications is very intolerant of any loss or any delay. For video, a
With individual public internet circuits at each office, you can offload delay greater than 200 ms is too long.
3 the demanding real-time video traffic from your own WAN by taking
advantage of the global, high-bandwidth, low-latency connectivity
offered by Lifesize and the IBM SoftLayer network

Video Network Best Practices | 4

Video Conferencing Network Best Practices

A Secure Foundation for Secure Video Calling

The early days of video conferencing required on-premises hardware that created a two-way
street of information going in and out of your private network, with inbound traffic potentially
opening you up to massive security issues. With Lifesize, we’ll never ask you to open inbound
ports from the public internet to your private network. We won’t ask you to alter your internet-
facing security posture, and we won’t make you spend more than you need to for security peace
of mind. Instead, our architecture allows you to keep your Lifesize room systems and client
software safely behind your firewall and manages firewall traversal through our global calling
nodes.

Lifesize room systems and client software do not require

any firewall ports to be opened inbound from the internet.
Detailed information about our compliance
There’s also no need for static public IP addressing or
certifications, in-meeting security features,
complicated static NAT (network address translation)
authentication and privacy policies can
and port-forwarding firewall configurations. Lifesize only
be found in our Lifesize Security Overview
makes use of outbound TCP/UDP connections for call
document. These are your top-secret
setup and media. These TCP/UDP connections are always
conversations we’re talking about, so we
initiated by the Lifesize endpoint or client software in
don’t take security lightly.
order to establish pinholes and dynamic port address
translations. These connections are always directed
to one of our global calling nodes on a specific list of
published host IP addresses allowing for tightly crafted
firewall rules.

Video Network Best Practices | 5

Video Conferencing Network Best Practices

Where the Network Ends

As IT pros, we can do our prep work, maximize our bandwidth and plumb all
our lines perfectly, but in the end, we must all come to terms with the fact
that our control has a defined limit. That’s why it’s so important to be able
to trust your provider to build a service that runs as efficiently as possible
to support the widest range of scenarios. From the teleworkers connecting
from their home offices to the traveling sales force logging in from the haziest
public WiFi networks around the globe, Lifesize is designed to support your
team collaboration on the broadest range of devices and network conditions.
Error concealment and correction were developed to help remedy network
imperfections, and audio-only options are available with local dial in numbers in
over 60 countries for truly poor networks unable to support a video feed.

For more information on topics like network readiness

and Lifesize cloud architecture, check out the Lifesize
Community. The Community is a place where thousands of
customers, partners and Lifesizers come together to share,
chat, geek and grok Lifesize. We can even arrange an onsite
visit to do a full network assessment and optimization.

Or, if you want to get a baseline for how Lifesize will perform
on your network, download a free 14-day trial of the Lifesize
App to try out on all of your devices.

HEADQUARTERS APAC REGIONAL OFFICE

Austin, Texas, USA Singapore
+1 512 347 9300 +65 6631 2831 www.lifesize.com
Toll Free US +1 877 543 3749 E-mail: [email protected]

EMEA REGIONAL OFFICE © 2017 Lifesize, Inc. All rights reserved. Information contained in this document is subject to
London change without notice. Lifesize and the Lifesize logo are registered trademarks of Lifesize,
+44 207 5962 835 Inc. All other trademarks are the property of their respective owners.

Video-Network-Best-Practices_0517