Monthly Archives: September 2014

You are browsing the site archives by month.

Lync Call Concurrency Calculator

Sizing appropriate hardware, bandwidth, channels, etc. is one of key activity task during the design phase of any UC project. One value which is extremely important and help you size correctly is the concurrency level of any modality – Voice, Video, desktop sharing, etc. If you are deploying a new solution you would need the concurrency levels to size up the new environment. If you are not deploying the new solution you still require the concurrency levels for reporting/billing/bandwidth forecasting purposes.

Surprisingly, I have found getting concurrency levels (e.g. concurrent voice calls) from existing environment is pretty hard. It is something pretty straightforward to work out but either the customer doesn’t have any logging tool on their ageing PBX or the PBX vendor just don’t want to give any details which may help with Lync deployment. Even after you have Lync deployed there isn’t anywhere in the Lync Monitoring reports to tell you what concurrency levels are for any of the modalities.

Concurrency Script

Over the years I have developed various tools/scripts which I use to gather information during the design phase. One of this is the Concurrency Script. This is an extremely simple but very powerful PowerShell script. This script allows you to generate reports of what your Call Concurrency levels are not just for Lync but from PBX data as well.

All that script requires is an input of a CSV file which has StartTime and EndTime columns which essentially shows the date and time when the call started and when the call ended. The script processes this information and returns you the peak load at every minute/hour/day/week.

As the script is looking for just StartTime and EndTime, the concurrency is not limited to calls, you can calculate the concurrency of anything….literally, even something not related to Lync, PBX or even Voice!

The logic of the script is simple. The diagram below illustrates this very well:
image

Read More →