It takes a lot of technology to blend together a unified communication platform. In this video, you’ll learn about unified communications and how VoIP, video streaming, presence, and QoS work together to make UC a reality.
<< Previous: High AvailabilityNext: Virtualization >>
In a very short period of time we’ve seen the way that we communicate change dramatically. We’re now able to communicate with voice, and video, and instant message, and desktop sharing. And it’s all happened in a way that has merged these together into something called Unified Communications.
There’s no exact formal definition of UC. It’s something that is really defining itself as we’re building it. But it generally is a way where we can receive a communication from someone in any form, and we can view that communication in any form. And all of these different communications can interact with each other in a way that’s easiest for us.
One technology that’s really driven Unified Communication is Voice over IP. This allows us to put our voice communication in IP packets, rather than using our older style, public switched telephone network, or PSTN. A federated version of Voice over IP means that there are no centralized exchange points, and that everybody can communicate to everyone else using this Voice over IP technologies.
We’ll generally see some very specific protocols used for Voice over IP. There are protocols to initiate a call for the call set up, there’s other protocols used to tear down the communication, and many, many other protocols are used to send the communication from one side to the other. Different providers of Voice over IP do things different ways, and you’ll find that there are a combination of standard protocols and proprietary protocols used all at the same time. And to complicate things further, different Voice over IP providers may be doing things in very different ways.
Another technology that’s important to Unified Communication is video streaming, the ability to send pictures over the internet, and moving video over the internet. There are generally two different ways to do this. One is through a unicast, where one device is sending information directly to another device. And if you have more than one user, there’s a completely separate stream of information being sent to all of the additional users. If there are three users, there’s three streams of video. If there are 20 users, there are 20 streams of video. Even if the video is identical, there’s a separate stream for every user.
You’ll often see inside of organizations a more efficient way of sending video through multicast, where the server can send one set of video out, and anybody who would like to view or subscribe to that multicast is able to watch it. That’s obviously a much more efficient way of sending the video, and it certainly scales up into much larger amounts of people. One of the challenges with that is that your equipment and your software have to be able to work properly in a multicast environment. We don’t generally see multicast being used on the internet. It’s usually something that’s used inside of an organization.
An important concept used in Unified Communication is the concept of presence. This is where all of these communication methods know what you’re doing at any particular time. They know if you’re available, they know if you’re driving down the street, they know if you are in a meeting. And the communication will adjust itself, depending on what your presence happens to be. You would make this status available to everyone, so that people can see what you’re doing in a moment.
Some of this presence may be calculated automatically. If you get in a car and you start driving, your mobile device may determine that you are driving now down the street, and can set your presence properly. Your applications then know how to contact you. If somebody is sending you a text message, you’re going to have that text message turned into speech, and communicated to you over the speakers in your car so that you don’t have to look down at any device.
It might look at your calendar to determine if you’re in a meeting, and not call you directly. Instead, take a voicemail convert that voicemail into text, so that when you’re in the meeting you can read the voicemail, instead of interrupting everybody by listening to it on your mobile phone.
From a networking perspective, we have to be concerned about all of these different protocols going over the network, and how they’ll affect all the other traffic that might be on the network. One way to manage this is through Quality of Service, or QoS, where we can prioritize certain types of traffic over others. We may want to have Voice over IP have a priority over someone who simply surfing the web. And that way you’re able to set bandwidth requirements, you can set traffic rates, you can assign separate VLANs, just to make sure that the important voice communication is able to get through, even if somebody else’s performing a file transfer or surfing the internet.
We described this prioritization as Quality of Service, or QoS. This is how we are configuring and setting up our systems to provide the prioritization of one application over another. And there’s a lot of different ways to do this. You’ll find that there are ways to do it inside of the applications, but there are also methods in the network protocols themselves that allow us to define QoS.
There are a number of different standards for Quality of Service inside of the network packets themselves. One is called Class of Service. This is inside the frame of the ethernet, and it’s part of the header. It’s in an 802.1 trunk, so this is something that you would be sending across a trunk connection. This is usually applied in your intranets, since you’re not generally trunking from an ISP. So this generally deals with traffic that is inside of your organization.
Another way to manage Quality of Service is through something called DiffServ. That stands for Differentiated Services. And we are setting bits to define Quality of Service in the IP header itself. These bits are separate to the application. We’re generally defining them and setting them up in the network equipment themselves, so the routers and the switches that we’re using have to be able to recognize those, and prioritize things appropriately.
Another method for defining QoS is also in the IP headers, using something called DSCP, or Differentiated Services Code Point. There’s a field in the IP header set aside for DS, or Differentiated Services, so that we can define how the traffic should flow through the network.
To be able to coordinate these Unified Communications, we need UC servers. And they are usually in our organization as a central point that is sending and receiving all of these points of communication. It’s going to connect users, whether they are sitting at their desk, or at home, or mobile, and it generally will also provide a way for third parties to interact with that UC environment.
Our UC devices these days generally take the form of a mobile phone. We might also have UC software that’s running on our desktop, so that we can easily migrate between one and the other, as we’re moving from our desk, moving into a mobile environment, and then back again.
If you have third parties that you’d like to interact with using this Unified Communication, you might need a UC gateway. This allows us to connect a lot of different systems together, and especially communicate to other UC systems that might be outside of our network.