Better Together: Microsoft Teams CQD and Synthetics Monitoring
Optimizing Teams call quality requires continuous testing, measurement, assessment, and evaluation
November 10, 2020

Sidharth Kumar
Exoprise

Share this

Maintaining call quality with Microsoft Teams is a process, not a onetime event. Network engineers and Microsoft Teams application owners need to be vigilant in preserving optimal call quality to ensure audio, video, and screen-sharing always remain satisfactory for end-users.

And vigilance is just as important before the pandemic as it is during the pandemic no matter where your users are working from. In some ways its more important today when working from home.

In this blog, we cover how the Microsoft Teams Call Quality Dashboard (CQD) combined with the audio/video synthetic transaction monitoring improves this maintenance process.



Teams Audio, Video and Messaging Synthetics

Once Microsoft Teams has been adopted for business collaboration, service owners will need a plan for ongoing monitoring and maintenance. At a minimum they will need to stay on top of the call quality with frequent visits to the Call Quality Dashboard. Proactive synthetics should be a part of everyone's Microsoft Teams monitoring strategy. Key, high-level benefits of proactive synthetic testing of Microsoft Teams include:

■ Network scenario what-if, testing and experimentation with no impact to the end-users

■ Faster network change management and root-cause capabilities

■ Immediate detection of outages before they impact end-users

■ Crowd-sourced benchmarks for networking jitter, network congestion, dropped packets and more

■ Network modeling of different topologies like SDWAN, VPN, Proxy, and Wi-Fi access points

By utilizing a proactive monitoring solution alongside the Teams CQD, professional network engineers can generate monotonic, Teams Audio and Video sessions which assist with diagnostics and Teams call quality maintenance. This improves productivity through multiple ITOps teams across multiple functions. Ultimately it improves business networking and productivity.


Better Together, Synthetics and the Call Quality Dashboard

Data for the CQD is collected in Near-Real-Time (NRT) and typically takes about 30-60 minutes before it is loaded. Synthetics can send data to its servers in real-time and contribute data to the CQD. This provides the best of both worlds; proactive monitoring from Synthetics with long term call classification from Microsoft in the Teams CQD.

A single sensor for monitoring Microsoft Teams Audio/Video streams can operate alongside any workload, physical or virtual computer. Monitoring can easily be deployed alongside a user's home or corporate machine if the home ISP or VPN connection is suspected.

Once deployed, Teams sensor will begin immediately collecting network and Teams telemetry, uploading it to monitoring servers in addition to the Call Quality Dashboard for data collection. The quality of experience (QoE) data that clients send to Microsoft 365 or Office 365 is the same from the synthetic Teams sensors. Importantly, this means no waiting weeks for enough real user samples to be generated for conference or peer-to-peer calls. Immediately benchmark and compare network telemetry. Immediately capture CQD data 24×7 for each sensor location and each network configuration.

Easy Quality of Service (QoS) What-If Scenarios

Quality of Service (QoS) for Microsoft Teams is critical for optimizing audio and video streams. When configured properly end-to-end in a network, QoS policy and settings enable network traffic from Microsoft Teams to be prioritized versus other traffic on the network. Voice, video, and screen share streams get dedicated ports of network bandwidth.

But Quality of Service configuration can be tricky to establish and ensure is operating well. Often new network components are introduced, settings lost or changed, and possibly new routing introduced. Ensuring QoS settings are properly established and maintained is another continuous process that is made easier by the Call Quality Dashboard and Exoprise together.

Armed with Teams synthetic sensors, network engineers can adjust QoS shaping and receive immediate feedback from analytics as well as Call Quality stream classification built into the CQD from Microsoft. Without synthetics, administrators are forced to wait for long periods of time to collect enough samples to ensure end-to-end QoS is in place. Additionally, without synthetics there is a potential to more severely impact end-users during Teams calls.

For QoS to be effective, changes need to be made consistently and throughout the organization because any part of the call or network path can degrade QoS policies and priorities. This includes applying settings to all user PCs or devices, network switches, routers, gateways, and proxies. Typically, these changes are made and coordinated across various teams which leaves more possibility of introducing errors. Customers typically experience savings throughout the year with QoS implementation.

Crowd-Sourced Network and Tenant Benchmarks

Wonder how your network or Teams user experience stacks up? With crowd source monitoring for Microsoft Teams, IT teams can instantly benchmark the Microsoft Teams digital experience and see how the network measures up against other enterprise customers.

The Value of Synthetics: Know Before That Important Teams Call

Knowing ahead of time about dropped calls, inability to connect with the Azure infrastructure or incidences of jitter, packet loss and high latency is critical to ensuring maintaining employee and business productivity. While the near-real-time collection and stream classification of the CQD is admirable, there are things that the CQD cannot provide:

■ Real time alarming for call failures and dropped calls

■ Alert notifications for poor network performance such as increased jitter, packet loss or Round-Trip Times (RTT)

■ Proactive notifications in advance of Microsoft Teams conference calls

■ Incident management integration with Microsoft Teams monitoring

■ Hop-by-hop contextual information about where in the network the problem exists

Service Availability from Every Vantage Point

While the CQD is great for stream classification it doesn't provide for availability metrics from each vantage point; branch office, network topology and user location. Utilizing synthetic Teams sensors from critical locations to capture accurate uptime and reliability statistics and compare to internal Service Level Objectives (SLOs) and Service Level Agreements.

Sidharth Kumar is Director of Product Marketing at Exoprise
Share this

The Latest

November 24, 2020

Shoppers are heading into Black Friday with high expectations for digital experiences and are only willing to experience a service interruption of five minutes or less to get the best deal, according to the 2020 Black Friday and Cyber Monday eCommerce Trends Study, from xMatters ...

November 23, 2020

Digital Experience Monitoring (DEM) has become significant to businesses more than ever. Global events like Covid continue to disrupt best practices within IT to support business. The pandemic has already forced millions of employees to WFH and adopt a hybrid workspace. Network connectivity and cloud application issues in these environments will continue to impact productivity and slow progress. Even so, transparent migration and deployment of on-premise workloads across multi-cloud providers, by their very nature are complex ...

November 20, 2020

APMdigest posed the following question to the IT Operations community: How should ITOps adapt to the new normal? In response, industry experts offered their best recommendations for how ITOps can adapt to this new remote work environment. Part 5, the final installment in the series, covers open source and emerging technologies ...

November 19, 2020

APMdigest posed the following question to the IT Operations community: How should ITOps adapt to the new normal? In response, industry experts offered their best recommendations for how ITOps can adapt to this new remote work environment. Part 4 covers monitoring and visibility ...

November 18, 2020

APMdigest posed the following question to the IT Operations community: How should ITOps adapt to the new normal? In response, industry experts offered their best recommendations for how ITOps can adapt to this new remote work environment. Part 3 covers automation ...

November 17, 2020

APMdigest posed the following question to the IT Operations community: How should ITOps adapt to the new normal? In response, industry experts offered their best recommendations for how ITOps can adapt to this new remote work environment. Part 2 covers communication and collaboration ...

November 16, 2020

The "New Normal" in IT — the fact that most IT Operations personnel work from home (WFH) today — is here to stay. What started out as a reaction to the COVID-19 pandemic is now a way of life. Many experts agree that IT teams will not be going back to the office any time soon, even if the public health concerns are abated. How should ITOPs adapt to the new normal? That is the question APMdigest posed to the IT industry. ITOps experts — from analysts and consultants to the top vendors — offer their best recommendations for how ITOps can react to this new environment ...

November 12, 2020

The pandemic effectively "shocked" enterprises into pushing the gas on tech initiatives that, on the one hand, support a more flexible, decentralized workforce, but that were by-and-large already on the roadmap, regardless of whether businesses had been planning to support widespread work-from-home or not ...

November 10, 2020

Maintaining call quality with Microsoft Teams is a process, not a onetime event. Network engineers and Microsoft Teams application owners need to be vigilant in preserving optimal call quality to ensure audio, video, and screen-sharing always remain satisfactory for end-users. In this blog, we cover how the Microsoft Teams Call Quality Dashboard (CQD) combined with the audio/video synthetic transaction monitoring improves this maintenance process ...

November 09, 2020

For IT teams, catching errors in applications before they become detrimental to a project is critical. Wouldn't it be nice if there was someone standing over your shoulder, letting you know exactly when, where, and what the issue is so you can correct it immediately? Luckily, there are both application performance management (APM) and application stability management (ASM) solutions available that can do this for you, flagging errors in both the deployment and development stages of applications, before they can create larger issues down the line ...