Big Data Analysis

Challenge

A leading telecom provider in India wanted to completely replace the tape archival process. We were engaged to create a low-cost, six-year CDR data pool archive consisting of Hadoop Distributed File System and object storage. 

Scope 

• Extract the CDR data from the billing server, 

• Stage and index the data and store it in low-cost S3 storage, 

• Develop search functionality for the archived data. 

Solution

• Implemented the solution using client’s Storage GRID object storage product enabling easy installation, efficient management and scaling of software-defined, object-based storage platforms at a lower cost, 

• Implemented Scalable High Performance Indexing, 

• Complex search, data analytics, reporting and interfaces for new algorithm development environment.

Outcome 

• Ability to store and process huge amount of data quickly, 

• Instantaneous CDR search and extraction for complex queries, 

• 1TB Call Detail Record (CDR) data generated per day, 

• Operational efficiency improved by five times, 

• Online access to six years’ data to develop analytics to improve customer lifetime value (CLV).