Get the video data you need to power Observability & Optimization
Solving Challenges With Streaming Video Observability
Making connections between data sets, coming from a myriad of sources, is critical to helping you optimize your service and network. With that kind of observability in your streaming video operations, you can meet, and even exceed, important KPIs to provide the best possible viewing experience.
01
Can You Easily Derive the Insights You Need To Make Better Business Decisions? How Would You Rate Your Streaming Video Observability?
Observability In Streaming Video: Seeing the Forest for the Trees
Data by itself can only reveal so much. It’s just the very surface of true analysis. And although visualized data can be used to draw obvious conclusions, such as a poorly-performing CDN, it’s much harder to just use the data as it comes to derive insights about business-related issues, such as grouping subscribers by their viewing habits and relating that to churn. This kind of analysis is called observability and is the keystone for organizations who strive to be “data-first.” But there are challenges to implement observability in streaming video and enable your operations engineers to see the forest for the trees.
Dataset Consistency
There may be times when data from different sources doesn’t use the same variable names. This can make it difficult to relate the datasets together without giving the variables a common name. For example, the concept of an ID (whether session or user) may be represented differently. These differences can make it very difficult to obtain any kind of observability.
Addressing this means:
- Using a standardized variable name across datasets
- Reducing the need to figure out data relationships
- Deriving insights more easily
Manually Relating Data
When data is inconsistent, it can require a lot of post-processing to compare datasets and identify the common variables which can be used to relate them. But this kind of manual effort takes a significant amount of time. When the insights that need to be derived are related to root-cause or performance, time is of the essence.
Addressing this means:
- Enabling employees to make quicker observations
- Ensuring variables across datasets match
02
Achieve Insight With Datazoom
How Datazoom Can Help
With the Datazoom Platform, you can tackle those observability challenges. By enriching and standardizing data gathered through various collectors, you can ensure the data which reaches your visualization tools or storage locations is already related. This can significantly increase the speed at which your network engineers, product owners, or business executives are able to derive the insights which can both protect and grow your revenue. Stop wasting time post-processing your data. Let Datazoom put together the data you need and deliver it, in real-time if needed, to the tools your teams already use.
Aligning with Standards
To improve data consistency, Datazoom has aligned its platform’s standardization feature with existing efforts from industry organizations:
- Common Media Client Data (CMCD). This standard is being developed by the CTA-WAVE group and addresses a common set of key/value pairs that players can communicate to CDNs about media.
- Common Media Service Descriptions (CMSD). This standard is being developed by the CTA-WAVE group.
- Distributed Request Tracing. This specification is being developed by the QoE/Measurement Working Group in the Streaming Video Alliance.
Arjun Patel
Network Engineer
Arjun is able to leverage Datazoom’s data standardization and enrichment features to create a single data set that he can have delivered in real-time to existing visualization tools like Datadog. The time saved post-processing data can be used by his team to identify new opportunities for viewer engagement and revenue.
