IBM®
Skip to main content
    Israel [change]    Terms of use
 
 
 
    Home    Products    Services & solutions    Support & downloads    My account    
IBM Research

Massive Collection System (MCS)

Business Transformation & Optimization


Telco Case Study


For many of today's industries, massive collection of data at high rates is critical. The telco industry is one such example, where the rate of network activity records may generate up to a billion call description records (CDRs) per day. In this industry, MCS targets mediation, enterprise data warehouses, fraud control systems, and revenue assurance systems.

Telco mediation - MCS processes CDRs and prepares them for post processing in the billing systems.

Telco revenue assurance - MCS collects all CDRs, enabling complete analysis and identification of revenue leaks by any revenue assurance application. Presently, some 6 - 12% of revenue is known to be lost in this industry. MCS supported revenue assurance drastically reduces loss of income and maximizes ROI within a very short period.

Teleco Revenue Assurance

MCS is currently being used for long-term revenue assurance at one of India's largest telcos, collecting and enabling a complete analysis of all CDRs. MCS formats the CDRs as unified ASCII records, filters and aggregates them, and then checks for duplicates and sequence gaps in records and in files. The records are then processed and uploaded to database tables. Intake is expected to cover 700M records a day.

Enrichment and business rules can be applied to the input records, based on the type of data being processed. The records are further managed in a DB2 database, using a partitioned architecture on a cluster of machines.

Subsequent processing is done using the Acsential ETL tool for dimension-based analysis of revenue discrepancies. This part of the system includes a dashboard application provided by Connectiva®, which enables users to drill-down to detailed CDRs stored in the MCS database tables.

All CDRs are kept online for a period of 1.5 months, requiring a total database storage of 100 TB.



 
 

    About IBMPrivacyContact