Streaming video operators face a continual enemy in the battle to ensure a high-quality viewing experience: root-cause analysis. Identifying and correcting the fundamental issues that undermine viewer QoE is a top priority, which is why many operators are turning to common media client data (CMCD): to get a better understanding of what’s happening in the player. However, the heart of most quality issues isn’t in the player, it’s in the content delivery network (CDN). Understanding those CDN issues, though, requires access to CDN data, which can be challenging to get in real time. Any delays in accessing this crucial data have long-term impacts, such as subscriber attrition. What many streaming operators don’t know is that they have the data to identify root-cause CDN issues in that same CMCD data they are using to better understand the player.
Why CDN log data is key to solving QoE and CDN issues
Most streaming operators have embraced multi-CDN delivery. According to the Bitmovin 7th Annual Developer Report, 47% of surveyed respondents were using multiple CDNs (where multiple includes DIY and third-party providers) while only 4% were not using a CDN at all. CDNs are clearly a vital part of streaming video architectures. CDN log data, when combined with player and other data from workflow components, is crucial to identifying root-cause analysis. However, if you’re using a CDN, and, more importantly, multiple CDNs, getting access to log data can be challenging. Plus, when every second counts in resolving QoE-related CDN issues, the delays in getting CDN log data can result in more viewers experiencing the problems that an operator is trying to resolve. Streaming operators desperately need a way to get the data they require, without adding time and complexity, to gain visibility into a CDN when it is involved in a QoE-related issue.
The challenges of getting to that CDN data to solve CDN issues
The difficulty of getting to the CDN log data quickly enough to help with root-cause analysis isn’t the only challenge. There are a number of other CDN issues that complicate using CDN data in root-cause analysis, all of which contribute to increasing MTTD and MTTR:
- Data ingestion: CDN logs can be terabytes of uncompressed data. Even if getting the data is easy, pulling it into analytics tools takes time
- Data normalisation: Although you might ingest CDN data quicker using programmatic methods (if the CDN supports it), you’ll still need to normalise the data against your player and other components so that the data can be used within existing dashboards. Depending on the amount of data, this can add more time to being able to analyse sessions across the entire data set
- Data coordination: If you are one of the providers employing multiple CDNs, you’ll need to get the data and normalise it for each provider. Each time you have to repeat the process, you add more time
Of course, root cause analysis doesn’t only happen once. It may happen multiple times throughout the day, and each time, those challenges can impact MTTD, MTTR, and, ultimately, viewer satisfaction.
How to use CMCD to prevent and resolve CDN issues efficiently
It’s clear that streaming operators need CDN log data. Root-cause analysis requires those logs to properly triangulate what is affecting viewer QoE. However, what if you didn’t need the logs themselves? That’s where CMCD can come to the rescue.
You can use CMCD data for more than just player session QoE analysis. By analysing the CMCD data from different dimensions, you can correlate QoE issues within the player to applicable CDNs.
Here are some of the CMCD data elements you can use to identify specific CDN issues:
| CMCD data element | How it can be used to identify CDN issues |
|
Encoded bitrate |
Displays the actual bitrate served in a session |
|
Buffer starvation |
If there were any buffering events |
|
Measured throughput |
Real throughput perceived from the player + end-user bandwidth |
|
Session ID |
Identifies a play |
| Startup |
If the object is needed urgently due to startup, seeking, or recovery after a buffer-empty event |
Building in the capabilities to leverage CMCD data like this can be challenging in its own right. Thankfully, Touchstream has done the heavy lifting for you. By employing the Touchstream platform, operators can utilise CMCD in this versatile way, providing visibility into CDNs and CDN issues for root-cause analysis without the challenges of getting the CDN log data.
Better telemetry means fewer headaches about CDN issues
The advantages of leveraging your CMCD implementation in this way are significant, with zero downsides. Not only will you decrease MTTD and MTTR, but speeding up access to CMCD data that can identify CDN issues will help ensure the best possible viewer experience. Think about it. You are already implementing CMCD for better player session visibility. Why not leverage that to reduce the complexity of your monitoring operations? You’ll remove the need to coordinate access to CDN log data from multiple providers, as well as avoid the headaches associated with ingesting, normalising, and processing those large log data files. The end result is the same: better QoE for happier subscribers.
Interested in how you can leverage Touchstream to improve root-cause analysis using CMCD data? Register for our free upcoming workshop here.
