Grid and Autonomic Computing Technologies

Grid computing technology underpins Acxiom's Customer Information Infrastructure (CII). CII is the engine that powers Acxiom's client solutions; it consists of three main components:

a simplified data cataloging process
a super-efficient grid-computing technology
a streamlined data mart extraction engine to enhance timely Customer Relationship Management (CRM) decision-making

CII offers a flexible, scalable new approach to on-demand access, management and updating of customer data for advanced analytics, and database marketing.

Acxiom’s approach to grid computing interconnects large numbers of inexpensive commodity processors (nodes), organized into enclaves called “hives.” The grid management system dynamically allocates hive nodes to particular applications in a way that balances the priority of individual applications with the overall throughput of the system.

One research goal would allow the grid management system to automatically respond and adjust to changing loads and component failures. This goal relates to autonomic computing research, which focuses on four areas: self-configuration, self-optimization, self-healing, and self-protection.

ALAR's research in this area centers on:

System management processes to automatically detect node overloading and failure, select a replacement node, and migrate applications and data to the new node with minimal interruption.
Methods and algorithms to automatically partition large datasets across tens and hundreds of nodes in ways that optimize storage use and data access speed.

« Back to Research