Media Provider Scores During the Big Game

When streaming media providers and broadcasters deliver major sporting events online, success means troubleshooting real and potential issues on the fly. To ensure a leading media provider successfully streamed one of the biggest annual sporting events to a record-breaking 124 million viewers, Akamai used TrafficPeak, its managed observability service, powered by Hydrolix. Real-time visibility into streaming data enabled Akamai to balance traffic across geographies and surpass viewers’ expectations for a smooth, reliable experience.

It’s tough to see and understand issues during live events. This media provider needed instant traffic insights to tackle performance and security problems before viewers suffered. In particular, it needed the ability to quickly collect, analyze, and visualize data — such as geography, city, state, Autonomous System Number (ASN), and server details — under peak traffic volumes, in real time, and in a single view.

The media provider turned to Akamai for help. To satisfy its requirements, Akamai needed to query CDN logs in seconds. But it took 20 minutes during peak traffic volumes to collect and analyze DataStream 2 data and Common Media Client Data — far too slow for this major event.

Instead of opting for a solution like ELK Stack, which wasn’t quick enough, Akamai relied on its TrafficPeak observability platform to guarantee the media provider satisfied streaming viewer expectations. TrafficPeak, with its intelligent data management, offered an easy-to-use solution and customizable dashboards that perfectly suited the media provider’s needs.

Running on Akamai Connected Cloud, TrafficPeak can ingest, query, and retain massive amounts of data for long periods at a fraction of the cost of other options. Plus, as a fully managed service, it cuts down on the resources needed to manage and analyze large data volumes.

After building out visualization around the top ASNs, response codes, and errors, Akamai was able to monitor delivery performance metrics, such as average bitrate. At peak traffic time, TrafficPeak collected nearly 11 million records per second, slicing and querying the data within subseconds. Real-time processing of massive data volumes enabled Akamai to swiftly pinpoint issues and their locations.

By instantly distinguishing provider-specific issues from those affecting ASN networks, Akamai quickly mitigated potential problems. For instance, when performance dropped in certain regions, it rerouted traffic before viewers were impacted. Plus, TrafficPeak provided real-time visibility into errors, enabling Akamai to address them on the spot.

In addition to gaining real-time visibility into game-day data, the media provider could store data affordably for post-game analysis. Overall, TrafficPeak collected 53 billion records and condensed 41 terabytes of raw data into less than 6 terabytes of compressed data for storage.

Using TrafficPeak for post-game analysis, Akamai drilled into performance outliers, identified their sources, and gained a clear understanding of their impact. Sharing these insights, Akamai empowered the media provider to preemptively mitigate such issues in future events.