Let’s take a glance at the Skype for Business Call Quality Database. This application was pretty easy to install. I generally followed the install written by Richard Brynteson. His link is here-
After the install, I was not able to bring up the CQD web page immediately. The error was something to the effect, that he cube had not cycled yet. That made perfect sense to me. The cube is basically a copy of the data from the monitoring/Archiving database. I decided to leave it alone for a few days. When I returned. All was working as expected and I got my web Page as expected.
Before going on, I will mention that some people had issues beginning with the application, The key statement I will make, is just authentication, time, and placement of your Databases as it relates to the over Lync Install. Lync has a lot of SQL going on. This is why it is recommended to create a server specifically this kind of data collection. I actually put mine on my Back end SQL server. With my whopping 2 users, I don’t anticipate a problem. I am in a Lab environment. What is your excuse? All I will say is your SAAS will take away from your lync/SFB deployment speed. Please put your bits on another machine for supportability reasons.
So the first thing you want to do once you have some data is to check your CQD health. Navigating to http://localhost/CQD/#/Health I see things look okay to me:
Now where do I go? Well It looks like the Blue bar at the top takes you back to the top of the slide. I think this is a top to down layout. So really, this is not as easy to navigate, but I think you won’t get to lost, as long as you understand to just go back to start, once you have drilled to the bottom of the stream you are in. For system health, there is no drill down, so “Skype Call Quality Dashboard” takes me back to the “home page” See Below, it doesn’t look too exciting so far.
The above Is a representation of a “monthly trend”. Ok well what does that mean. Let’s drill into that by clicking on the title of the report. The result is a breakdown of the graph into managed, unmanaged, and other. I assume, these are disjoint in terms of data streams, so we don’t want to have more complex reports, be tainted by inequivalent data points. So drilling into the managed audio, It really starts to break out into multiple paths. So at this point my path is:
All Audio Streams ->Managed vs Unmanaged Audio Streams-> Managed Audio Streams by Scenario-> Server-Server Dimensions
See the example graph below
In spending some time clicking around in the tool, What I found was that you basically start with one data set analysis, which is pre-defined. It is really not all that usefully just looking at what is pre-configured. What I found Is that the strength of the tool is the power to create additional graphs with the same data set. In other words you have the power to create graphs for things that your concerned about seeing. This is very good news, compared to the lack of this within the monitoring reports. Below find a setup screen when you choose new, under all audio Graphs.
In addition to custom reports, you can also stack the metrics on one graph. This may improve the amount of information the graph is relaying to the reader Below you can see the process of creating the custom graph with some metrics I have checked.
SO how does this Call quality dash board work? I say it works ok. It is a little slow in my lab. But Resources are also at a premium. I will say that I feel like it does not get the type of detail, in the way I would expect. I want to attribute this to a lack of large data in my lab. I am sure with a larger data set, these reports will come alive and be more meaningful. It is worth noting that when I did put multiple Metric Items on the report, It looked like it only showed one data metric. I don’t know if that is by design. However, I suppose I will leave it to you to decide for yourself. What do you thing of the new Call quality Dashboard?
Let me conclude by saying I think the Call Quality Dashboard is a good tool, that will help in collecting data and analyzing it until something better comes along. I know that Microsoft is really getting into Business Intelligence,
So I am sure more good things are just around the corner!