Exalytics incorporates an in-memory database, a design principle that will put it in direct competition with in-memory computing platform. In-memory databases place information to be processed in RAM, instead of reading it off disks, providing a performance boost.

The Exalytics machine includes 40 processor cores and 1TB of DRAM, but can hold five to 10TB of data in memory thanks to compression, Ellison said.

It runs a software stack that includes parallelized versions of Oracle's TimesTen in-memory database, BI (business intelligence) stack and Essbase OLAP (online analytical processing) server, Ellison revealed.

The system handles user queries "at the speed of thought" through a new interactive user interface, which will be demonstrated on Monday, Ellison said. The UI runs the same on PCs and the Apple iPad. "It really is quite spectacular."

An adaptive in-memory cache decides which information should be stored in memory based on the ongoing workload, he said. "If people keep asking the same questions over and over again, we keep the answers in memory so we don't have to compute them again." The cache is self-tuning, he added.