by David Loshin
Big Data analytics can hit unexpected performance bottlenecks. In-memory DBMS platforms can't address the issues of ad hoc analytic queries, and Hadoop processing itself can overload the network. By providing local data caches near the points of data access or processing, data replication solves the problem by:
• Providing extremely rapid access to data from multiple sources, even in a mixed workload
• Reducing the drag on multi-way joins for complex queries
• Accelerating reporting for faster analysis, review, and decision-making
David Loshin from Knowledge Integrity, Inc. explains how data replication optimizes Big Data analytics performance by rapidly cloning and propagating data sets to multiple targets for real-time response to different business analytics needs.
23rd May 2012