Datazoom releases data dictionary for CDN log streaming
October 28, 2020
Datazoom, the Real-time Video Data Platform, has announced its end-to-end data collection, data standardization, and routing service for streaming Content Delivery Network (CDN) log data. Datazoom’s CDN Data Dictionary has been defined in compliance with the Common Media Client Data (CTA-5004).
An important service that CDNs provide is efficient caching of video assets on behalf of content publishers so that videos load faster, and at a higher quality. As the streaming video industry matures, viewers have greater expectations of the viewing experience and lower tolerances when that experience fails. Content owners who are facing mounting pressure to deliver a flawless viewing experience, are looking for cleaner and faster ways to understand the performance of critical services, including the experience delivered by their CDN partners.
In response, CDNs are making investments to release new and improved log streaming services. Both Fastly’s Real-time Logs and Akamai’s DataStream offer log delivery to a variety of third party analytics and data warehouse tools. Recently, Amazon CloudFront announced a new feature to enable log delivery to Kinesis Data Streams. However, each CDN has built their logging systems differently, and thus gaining insights into performance using data from multiple CDNs has been an industry challenge.
With the release of Datazoom’s CDN Collector and CDN Data Dictionary content owners can easily analyze delivery performance across various CDNs in real-time. This solution aims to replace siloed analysis and time-intensive and offline processes to gather and clean CDN data.
The Datazoom CDN Data Dictionary represents common data elements that can be retrieved from multiple CDN distribution logs, and standardized to match the naming taxonomy. So far the Data Dictionary includes more than 20 data points. Datazoom also offers a Collectors for Video Players and a Video Player Data Dictionary with over 200 data points standardized for compliance with the Streaming Quality of Experience Events, Properties and Metrics (CTA-2066) spec. With cleaned and standardized datasets available in real-time, content owners are a step closer to having end-to-end observability of the video delivery workflow.
“Akamai recognizes the value in providing our customers with the ability to process CDN data alongside other data sets. That’s the rationale behind our DataStream product and the decision to open it up to endpoints like DataZoom,” said Jon Alexander, Senior Director, Media Product Management at Akamai. “Joining CDN data with player data helps correlate CDN performance to user experience and can facilitate more efficient diagnoses of the cause and impact of issues.”
The solution works with CDNs who can send data to Datazoom’s API Collector endpoint, and can pass data to any Connector that offers an API endpoint to receive third party data. Datazoom’s initial release includes adds both Fastly and Lumen (Formerly CenturyLink) to the Datazoom’s Collector ecosystem, with two other CDNs Collectors planned for release by the end of the year.
“There has been a long-standing frustration around understanding incident analysis. Although the CDN is part of the critical path, correlating CDN performance logs against downstream incidents experienced by end-users is something that today takes days to weeks. With this solution we can bring that down to seconds,” said Diane Strutner, Founder & CEO. “Arriving at this type of analysis faster will decrease recovery time for incidents and create an opportunity to implement prevention strategies, as well as more effectively provide feedback or share data directly with CDN vendors.”
Some of the benefits that customers will gain from using this solution include:
– Customers gain access to a standardized data set for input into Machine Learning, and real-time automation services.
– Customers can correlate against other critical data sets, like video player data, decrease time to identify root cause, and interface with other monitoring and alerting strategies.
-Customers can send CDN data to Connector destinations, including Google BiqQuery, Azure Blog Storage, AWS S3, AWS Kinesis, and Splunk.