IP Connections
UC: What you need to know
Predeployment testing is critical to ensure network performance is not impacted by new technologies.
by Stephen Brown
Unified communications
(UC) represent an appealing alternative to
traditional communication processes due to
potential savings. Developing a basic
knowledge of the technology and
understanding what needs to occur before
deployment is important, as well as what to
do once it has been implemented.
Typically, when
implementing a new communications
technology, such as voice over IP (VoIP),
many network managers take one of two
approaches. They may install new
technologies and then address performance
problems as they arise, or they anticipate
that the addition of the new application
will increase bandwidth needs, and upgrade
their bandwidth capacity.
Site surveys and
understanding the basic network environment
is vital to ensuring a successful UC
rollout. Conducting a site survey can
identify and eliminate many performance
problems before deployment. Proper
predeployment testing also allows the IT
staff to understand overall bandwidth demand
and application performance, and establish
benchmarks for acceptable network
performance. This knowledge is critical for
determining how the network will handle the
new UC traffic and identifying any changes
that need to be made to effectively support
communications.
An analysis tool that
tracks, stores and analyzes long-term
activity will define what is considered
normal for a particular environment. The
insights on network and application
performance gained from the initial site
survey and by continual monitoring of the
added UC traffic also helps in intelligently
configuring alarms on the monitoring tool to
alert staff when application performance
deviates from the norm.
When implementing any
communications application, ensuring
bandwidth availability through quality of
service (QoS) is an imperative. Failing to
implement QoS opens up the possibility of
interference from other applications on the
network–known as contention. Contention
leads to common performance problems,
including jitter and packet loss. Throwing
additional bandwidth at the problem does not
solve the issue.
Even a network with large
bandwidth capacity can have poor call
quality due to network contention. With QoS
monitoring tools, IT engineers are able to
set and outline limits to give individual
applications like VoIP the highest
precedence available, thus allowing the
application to function smoothly by
providing the appropriate amount of
bandwidth needed.
When monitoring QoS
issues, IT engineers should consider the
technology as a whole, rather than
concentrate on one particular aspect of it.
For instance, in the case of VoIP, they will
need to set precedence appropriately for all
connections that compose a call, including
call setup and tear-down. A common mistake
is to set highest precedence for the
conversation and neglect the other
components, leading to poor call quality.
Unprepared IT staffs may
not consider how UC applications can
increase the opportunities for sharing
information. These applications can open up
holes or flaws that allow users to
accidentally or maliciously share private
data and expose sensitive information. These
new holes can also be points of attack if
not properly secured.
In the case of VoIP,
hackers or users can use tools to not only
capture but play back VoIP conversations.
Higher-end VoIP systems may offer ways to
encrypt the data, but lower-end products
often do not. In addition, VoIP traffic is
usually most vulnerable on the LAN since
Internet WAN traffic is typically routed
through VPNs.
To ensure security, every
communication channel should be properly
secured and monitored. Acceptable use
policies should also forbid or provide
protocols for the proper sharing of
sensitive information.
Stephen Brown is product marketing manager for network-analysis vendor
Network Instruments, Minneapolis.
For more information
(click here)