Skype for Business Reporting, Diagnostics, Data Modeling and how to leverage BI for Skype for Business

 

Good day all. I spent some time this weekend with the Skype For Business call quality metrics. In fact I am working with a team to try and come up with a battery of tests to fulfill the needs of different departments, In this area. The Idea is to develop and run a series of scripts, create some reports and follow a process to get a good snapshot on weather there is an issue in the environment. We want to know  if the network is acceptable for Skype for Business, and in general, is the customer at a minimum threshold of acceptability for an existing or greenfield deployment, or is work needed to bring them to this level.

This turns out to be a little difficult to quantify. We all know this is like herding cats to build something for end to end telemetry. I do not write this article as satisfied with my conclusion, but I do say I am encouraged. Let me say right up front, I did not find or develop that single tool to get this benchmarking accomplished. What I did find is the industry had identified this problem and has been working to fix this problem for the Industry at a whole. This means Skype for Business will be a direct benefactor of Microsoft posturing as a Telemetry and Business Intelligence Behemoth.

 

Below,take a look at company’s Microsoft purchased in 2015. This is a print from the Wikipedia search for “Microsoft purchases 2015”. If you notice, the number of Analysis, Predictive and otherwise, Data centric companies are a clear goal for Microsoft. Some of the fruit if this focus has ben realized already. The Call Quality Dashboard is one tool that is a step in the right direction. This came right at a time where people were feeling the burn of the difficulty, or should I say complexity of Lync problems.

5-15-2016 7-47-14 PM

While the Call Quality dashboard is just a step, there are several other example of Microsoft Filling in these gaps. Here are a few of the related tools:

1. Call Quality Dashboard

2. Key Health Indicators

3. Skype for Business (SFB) Monitoring

4. SDN (IR is in the example below)

I added in SDN in here because I think this technology represents why are  focused on telemetry going forward. I am not going into the “why” in todays discussion. But you can get an understanding of why Business Intelligence, Telemetry and instrumentation are moving us toward real time root cause and fixes. You have to register for the webinar, but check out the IR.com video. They have an SDN solution to check out.

The new application from Microsoft is called Power BI Desktop. This tool is not SFB centric, but is really a front end to do your own Data Mining. It is the most fantastic free BI tool I have seen yet. It represents the ultimate in configurability and customization. I will admit, I am still a little overwhelmed with it, But as I learn and use it more, I will certainly contribute what I can to its propagation and use.

 

The Business Inteligence and Data Mining are getting into more areas every day. My Firefox browser has a BI add-on to tell me who I am monitoring on the web, as well as who is monitoring me. It is all about how we use these tools, that will make use more and more successful. See the  Light Beam Add on below as an example of BI everywhere:

 

5-15-2016 10-01-54 PM

Finally, Let me talk a little about Skype for Business and Power BI Desktop.

I am in the early stages of trying to create some durable methodology. I began by taking a snapshot of the SFB QOE Metrics Database. I pulled it from SQL and saved it as a PBIX file. Now I can open this file and create reports at will. The thing that quickly overwhelmed me was the large number of tables, with little to no understanding of what was inside the tables. So I stated with a couple representative tables:

FROM [QoEMetrics].[dbo].[AudioStream]

FROM [QoEMetrics].[dbo].[CallSummary]

I just right click the table and selected “select top 1000 rows”. This gave me a basic query to work off of. Now I move to the BI desktop.

 

Then when I was ready to log into the Power BI, I put in the Database I wanted to connect to :

 

5-15-2016 11-08-48 PM

Once I connected, I got drowsy quickly,  browsing through all the tables, just like I was looking at it in SQL server. So I highlighted all the tables and just saved them. I selected all the check boxes for each table and chose load. once loaded. I simply went to the top left and selected File-> Save As->MYFILE. then I took MYFILE and put it on my desktop machine and searched for a few tables. you don’t use brackets, just enter the table you want, and it will come up with its value fields.

Your BI search is not centered on tabled however. you can only make operations to a data point. This is the field or columns inside the table. You can drag any element of a table onto the white desktop, and you get a graph instantly. See below I chose Degradation and Jitter interval.

 

5-15-2016 11-27-47 PM

 

When I click on the see records icon, I get the following:

5-15-2016 11-34-13 PM

So far so good. I must level with you. It has been several hours on this project today. There is a lot to learn about this tool and I don’t pretend to know anything about it.

In looking at just the paltry graphs I made, I could already see fields and information I had not been able to get easily before. Right in front of me, I can see the relay addresses, and the caller and callees subnet. Right away, I can start seeing patterns the may narrow a call quality issue to a particular subnet.

The technology is a little difficult to wrap your head around at first. At least for me it was and still is to a degree.  It started to make more sense as I used simple SQL queries to get table results. Then I compared that to what I was doing on Power BI, and looked to get the same result. The big difference being that the Power BI results have graphs, and then you can drill down into the graph. I give this a 10 compared to the call quality Dashboard. Its not that the call quality dashboard is bad, Its just the flexibility is so good on Power BI. Skype for Business is a good fit for Power BI Desktop.

Others think the same I believe There is a recommendation from users to port the current excel monitoring from Excel to Power BI. Read about it here

 

5-16-2016 12-05-01 AM

 

 

Send me any templates you find, and I will pass on any that I create!

Good luck and Skype on!

 

Louis

Advertisements

2 thoughts on “Skype for Business Reporting, Diagnostics, Data Modeling and how to leverage BI for Skype for Business

  1. Chris says:

    You probably missed the biggest step Microsoft has taken in moving their monitoring and analytics for S4B to the next level. In January Microsoft purchased the IP for Event Zero’s tools, which are widely regards as the best in the market. They announced this tool will be coming for O365 later this year.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s