Deploying an Elastic Data Fabric with Caché

Executive Summary

For twenty years or more, large financial institutions have been locked in a battle between the need for extremely high performance transaction processing and the demands of downstream applications that can deliver competitive advantage if they can get real time access to this transactional data. When individual database servers could no longer handle simultaneous transaction and query workloads, many firms turned to replication, offloading data access onto read-only copies of production databases. While this strategy worked well for smaller volumes, greater data volumes (years of accelerating trading velocity, more data sources) have stretched data replication architectures to the breaking point. Many are exhibiting problems with performance, scalability, manageability, maintainability, and data governance. In short, they have become an unacceptable risk to firms’ continued growth and market responsiveness.

InterSystems Caché offers a different approach, by enabling customers to deploy an elastic Data Fabric that is massively scalable, runs on commodity architectures, and can be deployed in the cloud. Instead of replication, Caché uses its advanced enterprise cache protocol that transparently delivers in-memory access speeds for massive amounts of data distributed in local- and wide-area configurations. To oversimplify, Caché replaces hordes of database replicas and database servers with a distributed, shared data cache designed for fault tolerance, data integrity, and linear scalability.

Most importantly, Caché includes robust SQL capabilities so that existing SQL-based applications can use the elastic Data Fabric without disruption. And, because Caché natively delivers other powerful data paradigms – objects, documents, key/value pairs, and so on – the elastic Data Fabric also transparently enables future transactional and analytic development, even with Big Data, on an enterprise-grade massively scalable foundation.