Stream Monitor - Overview

Overview

The Telestream Cloud Stream Monitor service deploys stream monitoring where and when needed. There are two primary types of monitoring enabled with the service:

• Contribution Feed Monitoring
• Edge Stream Monitoring

In many cases, a cloud contribution feed is transcoded and packaged for ABR streaming applications. The Stream Monitor service supports modeling a channel, consisting of the contribution feed and resultant ABR assets in to a single object that can present end-to-end correlated performance data.

Additional supported workflows include the ability to monitor re-distribution and streaming only workflows.

This diagram shows all of the locations you can now deploy monitoring and start producing realtime metrics for the streaming channel...

Projects > Channels > Assets > Streams

Stream Monitor organizes what's being monitored in to convenient objects that make it easy to quickly identify and correlate the health of your content delivery. The object hierarchy is Projects contain Channels which contain Assets which contain Streams. An example of this hierarchy...

Projects
When configuring Stream Monitor, the highest level object you'll need to configure is the project. Projects are simply convenient ways to organize collections of channels. For example, an organization focused on streaming several linear channels and frequent event channels, they may wish to have a linear project and a separate event project. There are a few settings that are applied at the project level, such as data retention period and ABR QoS Measurement & Report level.

Channels
Once you've configured a project, you can start adding channels. A channel can include one or more contribution feeds and one or more ABR edge streams. For example, an organization is delivering a contribution feed to a cloud live video transcoding and packaging service using SRT protocol. The cloud service is taking the contribution feed in and producing a live HLS and DASH stream from it. In this scenario, all of these assets can be part of a channel. By organizing related assets in to a channel, you can more easily identify and correlate performance and availability issues that may be impacting downstream playback.

Assets
Assets are where you define the nature of the content you wish to monitor. Assets include the type of stream to be monitored, including what protocol or package format is being used and how to connect to the asset. Assets are generally divided in to "Contribution Feeds" and "Edge Streams", where the contribution feeds are in the form of transport streams and the edge streams are in the form of http adaptive streaming formats. Contribution feeds support Zixi, SRT and UDP delivery. Edge ABR streams support HLS, MPEG DASH, Smooth and HDS package formats.

Streams
Asset will carry discreet streams. In the contribution feeds, this will be a 1:1 relationship with the asset. In the edge stream assets, the variants are represented as discreet streams. So, for example, a single HLS asset might include up to 20 variants, each one being its own streams in Stream Monitor.

You can see the assets and their streams within a single channel in this screenshot.

Performance Data, Alarms, Notifications and Reports

Stream Monitor collects a wide range of metrics and enables a broad set of alerts to help identify and remediate issues that may be impacting your ability to deliver a quality stream. In general, metrics are broken in to two categories...

QOS
QOS metrics measure the availability and overall network delivery performance for a selected asset. Stream Monitor presents a wide range of QOS metrics tailored to the specific type of asset being monitored. You will see QOS metrics specific to the contribution feed protocol (like Zixi or SRT), the health of streams as they are received in to cloud processing & re-distribution centers and finally CDN performance metrics measured on the content origin and all the way out at the CDN edge. By correlating these metrics, you can rapidly identify faults in the delivery network or availability issues that negatively impact delivery performance.

QOE
QOE metrics dig a little deeper in to the content itself, looking to measure audio/video quality, encoding performance and identify impairments that can lead to frozen video, encoding artifacts and silent audio.

Thresholds can be defined for most of the collected metrics in the form of alarm templates. It is easy to create alarm templates tailored to the content and expected network delivery performance. When QOS or QOE metrics are measured to be outside of the defined acceptable thresholds, alarms will automatically begin firing. Alarms are visualized in the realtime dashboards, logged in the event viewer and API endpoints and can trigger programmatic notifications. You can see an example alarm template for contribution feed monitoring here: