Case Study: Kcell builds a real-time stream analytics platform with Ververica and Apache Flink

How Kcell and GetInData built a stream analytics platform using Apache Flink® as a core component

Kcell, a major telecommunications provider in Kazakhstan, needed a way to react instantly to massive streams of billing, usage, roaming, and location data. Working with Ververica’s Apache Flink technology through GetInData, the company aimed to build a real-time stream analytics platform that could support use cases such as balance top-up offers and fraud detection while roaming.

Ververica’s Apache Flink-based solution, combined with Apache Kafka and Apache NiFi, enabled fault-tolerant, low-latency complex event processing with event-time accuracy and flexible data pipelines. The platform was designed to handle more than 10 million subscribers, around 160,000 events per second on average and 300,000 at peak, processing about 23TB of data per month while supporting 24/7 reliable operation, monitoring, and end-to-end testing for future growth.

Kcell