Services: Cloud Configuration Services
Big Data Volumes
Big Data Volumes for Telecom
Telecom is one of the biggest data consuming industries. Using Kafka-based real-time data processing systems can provide Telco’s tremendous benefits in terms of better decision making and revenue generation.
For example, from the call detail records (CDRs) alone, huge volumes of data are transacted every day. A Telco needs to process these CDRs real-time on a day-to-day basis.
A Kafka-based system allows:
-
The CDR files received from the mobile tower is streamed record by record to Kafka.
-
Aggregator application aggregates the CDRs based on the MSISDN for Call durations and Mobile Data consumptions.
-
Aggregated data CDRs are then streamed back to Kafka.
-
It is then stored in Elastic Search.
-
Aggregated reports are created and visualised in a Kibana dashboard.
Kafka-based System for CDR Aggregation
Technical Insight
Telco Call Data Record Aggregation with huge volumes near real-time:
-
To simulate the mobile tower, CDR files are generated by a Spring Boot microservice application and then they are streamed record by record to Kafka using the CDR Kafka Producer.
-
Microservice aggregator application written using Apache Flink aggregates the CDRs based on the MSISDN for Call durations and Mobile Data consumptions.
-
Aggregated data CDRs is then streamed back to Kafka using Apache Flink producer.
-
Kafka Connector is used to read the CDR aggregated data and update the data in Elastic Search.
-
The aggregated reports are created and visualised in a Kibana dashboard.
