Without Conviva, identifying and resolving these issues in real time would be impossible
Partnering with one of the world’s largest streaming media providers on streaming-exclusive NFL games, Conviva ensured the delivery of a seamless experience for football viewers by computing and troubleshooting user-perceived quality of experience (QoE) in real time. During one NFL game in particular, the Conviva platform processed 3.6 billion events per minute and 7 billion stateful metrics per minute, allowing one of the largest media streaming providers in the world to monitor the experience of more than 16 million users and react in real-time when issues with user experience occurred.
Data streaming technology used: Conviva Time-State for stateful real-time streaming, Apache Kafka, Apache Druid
What problem were they looking to solve with Data Streaming Technology?
In partnership with one of the largest North American streaming providers, Conviva was tasked to ensure the delivery of a seamless viewing experience for a series of streaming-exclusive football games. To effectively measure and monitor real-time viewer experience, Conviva activated real-time data streaming solutions, powered by our proprietary Time-State technology, to compute and troubleshoot video quality of experience (QoE), playback failures, and subscription/paywall issues, in real time. Without Conviva, identifying and resolving these issues in real time and at scale would be impossible.
How did they solve the problem?
Conviva tackled 3 problems in order to deliver a seamless viewer experience during these high-stakes, large-scale streaming events:
1. Full Census Experience Data Processing: Ingesting, normalizing and sessionizing high-volume viewer experience and system performance data to enable real-time monitoring for every viewer.
2. Real-time Stateful Metric Computation: Computing user-perceived quality of experience (QoE) over billions of events per minute and creating stateful experience metrics to quantify true user experience.
3. Real-time AI Alerts: Enabling automatic identification of granular user cohorts (e.g.. geo, ISP, devices) experiencing problems across millions of sessions, including explainable root-cause analysis to help operations teams to detect and resolve issues as they’re happening
What was the positive Outcome?
During the games, Conviva processed 3.6 billion events per minute and 7 billion stateful metrics per minute, allowing teams to monitor the experience of more than 16 million concurrent users. Additionally, Conviva provided these teams with a shared ‘common language,’ rooted in their users’ experience. Conviva enabled the teams to set and monitor KPIs for business growth, continuous improvement, and rapid response to user facing issues.