Skype for Business has an uptime guarantee of 99.9%, which is just about as good as you can ask for in this era of SaaS and cloud apps, right? Dealing with about 8 hours of downtime a year doesn’t seem terrible.
However, designing your Skype for Business deployment specifically to meet the minimum network requirements to achieve uptime isn’t the only thing that counts. When Skype for Business is a critical application for your business—for example, if your director of sales is trying to close a big deal—there are other metrics that become equally important. That’s why companies must zoom out to a bird’s-eye network view to get a comprehensive look at what contributes to the quality of the performance of applications like Skype for Business.
Meeting the Minimum Network Performance Requirements
To ensure excellent, not just good, real-time media quality, there must be sufficient bandwidth over the entire path that connects the network to Skype for Business Online.
When we talk about quality with a voice application, there are three influential factors:
- Latency: Network propagation delay from point A to point B on the network.
- Packet Loss: Percentage of lost packets in a specified period of time that can cause audio cutout.
- Jitter: Average change in delay between successive packets.
To optimize Skype for Business media quality, Microsoft suggests the following network performance metrics targets or thresholds:
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 |
Achieving and maintaining these numbers might be difficult for some, which is why Microsoft provides planning tools, and suggests Quality of Service (QoS) enforcement so that packets can travel through the network faster.
Why You Should Go Beyond the Threshold for Maintained QoS
Skype real-time media travels through a variety of devices, including servers and client applications, and across different networks. With the complexity of these networks and travel, it’s insufficient to simply meet the thresholds; instead, companies should strive to surpass minimum efforts to maintain quality of the end-to-end connection. This means actively working to achieve monitoring that can detail metrics such as those provided by a network router on a hop-by-hop basis.
Poor user experience isn’t just a vague possibility. Organizations that deploy Skype with the softphone application oftentimes find they are still struggling with voice quality. Given that it is such a highly visible and critical service, network teams must focus on stopping issues such as latency and jitter before users experience service degradation.
Traditional VoIP monitoring techniques that rely solely on call detail records are mostly reactive, which will almost guarantee unhappy user experiences. As the shift to cloud-based UC and VoIP continues, this will be highlighted, as quality over the Internet is generally considered to be “best effort.”
If companies want to be proactive in their Skype for Business quality, they must adopt a critical monitoring approach to include:
- Active monitoring of voice traffic that tests the VoIP call setup and the quality of the stream, from SIP signaling to RTP streams.
- Testing that can correlate SIP transactions, voice quality and QoS enforcement metrics to underlying network performance.
- Continuous measuring of network capacity between call source and destination to identify drops in performance during peak utilization times.
- Drill-down features by users, hosts, applications or conversations for faster remediation.
What to Do When Your Skype Voice Quality Isn’t Acceptable
Organizations that run Skype clients on laptops or PCs just can’t manage voice traffic in the same way as traditional voice deployments. Segmenting and prioritizing voice traffic over other business-critical data traffic is nearly impossible, but that doesn’t mean everyone has to suffer from congestion in silence.
AppNeta has developed a monitoring solution that can answer the question of “Is my network robust enough?” With continuous monitoring that is proactive by nature, calls traversing the network, including the WAN and third-party networks, can be overseen for issues specific to Skype.
To learn more about monitoring business-critical SaaS applications like Skype for Business, download our white paper.
Filed Under: cloud computing, Performance Monitoring
Tags: SaaS applications , SaaS monitoring , Skype for Business