IBM is releasing into the commercial arena the underlying storage technology that made its supercomputer Watson such a big hit on the Jeopardy TV quiz show, in which it trounced two of the former game show champions.
The technology is being released as a portfolio of software storage products designed to enable organisations to access and process any type of data. Big Blue claims users can carry out data processing out on any storage device anywhere in the world.
Watson processed 200m pages of information, including the full text of Wikipedia during its appearance. With each question the system scoured four terabytes of content to present the most probable response within a few seconds. About five terabytes of Watson’s knowledge were loaded in minutes into the computer’s memory during the show.
The Watson system had architectural limits which stretched into the thousands of yottabytes. A yottabyte is one billion petabytes, or the equivalent of a data centre the size of one million US city blocks.
One of the Watson portfolio technologies, codenamed 'Elastic Storage', offers infinite scale, and is capable of reducing storage costs by up to 90% by automatically moving data onto the most economical storage device, claims IBM.
IBM Research says it has demonstrated that Elastic Storage can scan 10 billion files on a single cluster in 43 minutes. IBM is pinning its hopes on the technology providing a shot in the arm for big data analytics.
Tom Rosamilia, senior vice president, IBM Systems and Technology Group said: “Digital information is growing at such a rapid rate and in such dramatic volumes that traditional storage systems used to house and manage it will eventually run out of runway. Our technology offers the advances in speed, scalability and cost savings that clients require to operate in a world where data is the basis of competitive advantage.”
Elastic Storage virtualizes the storage allowing multiple systems and applications to share common pools of storage. This enables global access to data without the need to modify applications and without the need for storage management applications.
IBM says because Elastic Storage is not reliant on centralized management to determine file location and placement, customers can have continuous and highly-available access to data in the event of software or hardware failures.