Project overview

A scalable QoS/QoE monitoring system that collects vast volumes of statistics for online TV channels, programming, and audience distribution by CDN, country, device, and the number of users. The solution supports historical and real-time stats and works smoothly with tens of millions of database records.

Client: a US-based OTT TV platform with a global audience and a focus on sports and entertainment programming.

Delivery Model
SLA-driven task backlog delivery (TM-based)
Effort and Duration
6 months, 30 man-months
Java, Spring, Hibernate, Apache Hadoop, Apache Kafka, MapReduce, MySQL, HTML5, CSS3, jQuery, Bootstrap

Business challenge

The owner of a popular OTT TV platform was looking to develop a next-generation monitoring solution to glean more real-time insights on the performance of the critical nodes in the video delivery chain.

The client was also interested in gathering detailed user behavior and user distribution data to make fact-based decisions about the platform’s future in terms of content acquisition, monetization strategy, and territorial expansion.

Delivered solution

Oxagile developed an OTT platform monitoring system that supports two types of stats: historical data used to understand utilization and solve infrastructure problems, and real-time data that enables OVP/OTT clients to effectively monetize viral videos and cover new territories.

Key features and highlights

  • Historical and real-time statistics
  • Ability to seamlessly support tens of millions of records
  • Horizontally scalable architecture that allows monitoring an unlimited number of users
  • Powerful report generation module that enables stakeholders to evaluate every aspect of the business, spot emerging trends, and allocate funds for maximum gain
  • Sophisticated user behavior analysis
  • Analysis and adaptation of video quality for multiple users in real time
  • Audience distribution filtering by CDN, country, device, and number of users
  • A convenient dashboard with the number and location of concurrent users, top-rated programs, QoS metrics for multiple live streams, and more

Project results

By harnessing a cutting-edge big data stack and pairing it with in-house OTT domain expertise, Oxagile’s team delivered a feature-rich solution capable of handling vast amounts of data from many distributed sources.

The insights provided by the system helped the client to implement the right multi-CDN strategy which resulted in a 47% drop in expenses on the US market alone in the first three months after launch. Furthermore, the client leveraged audience distribution analysis to segment ad campaigns for different device types, achieving a 58% boost in CTR.