MCS - Massive Collection System - overview
The IBM Massive Collection System (MCS) can process billions of data records each day and maintain them in a repository for drill-down. The system delivers continuous processing with full support for high availability, data recovery, and transaction-based operation. MCS is an enterprise-grade product based on the IBM Websphere(R) highly scalable Java 2 Enterprise Edition (J2EE) platform.
MCS begins by reading, in real-time and in batch, input records consisting of fields in multiple formats. It then goes on to parse, process and transform the data, enrich the data using lookup tables, detect duplicates and sequence gaps, filter the input, and produce aggregated and non aggregated output records. Next, the system uploads the resulting records to a relational database for easy access and retrieval. An innovative repository solution is Dynload, reducing database holding space by a large factor to an economically small size which is superior to any other available solution in the market.