SAP Speeds Up Business Intelligence with In-Memory Analytics
Updated · Dec 02, 2010
SAP HANA is bundled with hardware from partners IBM, HP, Fujitsu-Siemens, Intel, Cisco and Dell. The goal of the appliance is to process transactional data for real-time analytics.
With HANA, analytic applications “will run directly on top of large volumes of transactional data — offering customers instant visibility into their businesses,” according to Jacob Klein, head of the HANA project.
“What we’re doing with HANA is starting with an in-memory database and expanding that to provide deeper integration with the application server so that it will be possible to build more sophisticated, industry-specific applications that are designed natively for in-memory, beyond business intelligence scenarios,” Klein said in an SAP posting.
Those applications will include sales pipeline forecasting, smart meter analytics for utilities and consumer packaged goods and retail applications such as promotion planning.
On a conference call today, SAP executive board member Vishal Sikka discussed one retail application in which 460 billion point of sale records were queried in less than a minute on ten 32-core blade servers costing $530,000.
Sikka called in-memory technology a “once in a generation technology shift.”
The first application is the BusinessObjects Strategic Workforce Planning application, designed to process “large quantities of data using in-memory computing with an innovative calculation engine that enables application processing to be performed directly in-memory,” SAP said in a press release.
With the Strategic Workforce Planning application, managers can “simulate scenarios in real time, analyze complex relationships quickly and see how proposed organizational changes will impact the business,” SAP said. Managers and human resources executives can simulate how their workforce would need to grow and change using predictive modeling, allowing for data-based decisions about allocating or adjusting staff as needs change.
SAP claims to have “mastered massive parallelism for enterprise applications (such as aging) with its in-memory computing engine.” SAP HANA scales linearly, with performance proportional to hardware improvements, so “future customers will be able to deploy systems with 1,000 or more cores that enable complex real-time analytics.”
Paul Ferrill has been writing for over 15 years about computers and network technology. He holds a BS in Electrical Engineering as well as a MS in Electrical Engineering. He is a regular contributor to the computer trade press. He has a specialization in complex data analysis and storage. He has written hundreds of articles and two books for various outlets over the years. His articles have appeared in Enterprise Apps Today and InfoWorld, Network World, PC Magazine, Forbes, and many other publications.