GSX Blog

How monitoring Skype for Business Online can improve end-user experience

Posted by Pavan Reddy on Mon, Feb 06, 2017

SkypeAudio.jpgMany organizations are migrating to Skype for Business Online, and for good reason. Migrating to the cloud takes away the need to buy hardware for servers and worry about upgrades and maintenance. Companies can save money on backup and disaster recovery as well.

Most organizations think they don’t have to monitor their Skype for Business Online environments at all, since Microsoft promises 99.9 percent availability for tenants hosted on the cloud. It’s true that you don’t need any infrastructure to maintain the system, but it is still important to deliver the best end-user experience within your network.

So, how do you track end-user performance issues from different locations? There are multiple parameters in your internal network that can cause many performance issues for end-users. As you know, the quality of audio, video, and sharing over the IP is completely dependent on end-to-end network connectivity. It’s essential to understand the connectivity between internal and cloud network to fix these issues proactively.

Network Performance requirements from a Skype for Business client to Microsoft Network Edge (Microsoft recommended)

Metric

Target

Latency (one way)

< 50ms

Latency (RTT or Round-trip Time)

< 100ms

Burst packet loss

<10% during any 200ms interval

Packet loss

<1% during any 15s interval

Packet inter-arrival Jitter

<30ms during any 15s interval

Packet reorder

<0.05% out-of-order packets

How GSX will help the administrators to find out voice quality?

The GSX Robot User, a small piece of code that easily installed and deployed, can help administrators test the call quality on Skype for Business Online by conducting synthetic tests. The Robot User tests parameters including Jitter, Latency, Packet Loss, and MOS to deliver real-time information identifying issues with call and voice quality:

  • % Packet loss: GSX Robot User will request administrators to provide the Microsoft edge endpoint , if they are not aware by default will try to reach global anycast IP endpoint and get the complete details on the number of packets sent /received with the single scan and get  calculate the % of packet loss.
    If % Packet loss is more than 1% it definitely impact audio quality. GSX will allow administrator to threshold on % Packet loss and alert them in case the threshold exceeded the limit.

packet loss.png

  • Latency:  GSX will continuously try to reach the endpoint and detect the hops between the locations to the endpoint. GSX identifies the delay between the hops and immediately alert the administrators to fix the latency issues before end user escalated the issue.

network.png

  1. Jitter: Jitter is nothing but a delay of received packets. This can happen due to network congestion or configuration errors. By using Robot configuration, GSX will get the packet loss information and calculate the jitter from the Robot location and send the alert to administrators if the jitter threshold is more than 20MS. 
  2. Network MOS:  The MOS indicates the perceived voice quality of a VoIP conversation, ranking the call quality as a number in the range 1 to 5. GSX will analyze the MOS based on Packet loss, Latency and jitter from the location and provide the exact MOS result which helps administrators to proactively check the voice quality issues and fix it immediately.

Migrating to Skype for Business Online can ease many of the concerns and inconveniences faced with on-premise services, but that doesn’t mean you’re totally in the clear. With help from the GSX Robot User, you can perform tests from different locations, comparing the results to get the best possible understanding of your internal network issues before they impact the end-user on Skype.

CTA-bas--de-blog-1.png

Tags: Office 365 monitoring, skype for business issue, skype for business monitoring, SysAdmin Top Issues, office 365 health, office 365 performance, fixing office 365