IBM Debuts Elastic Storage Platform for Big Data

By Pedro Hernandez

IBM is commercializing some of the storage tech that helped Watson win big against the human competition at Jeopardy.

The IT giant today announced IBM Elastic Storage, a highly-scalable storage software platform based on the company's General Parallel File System (GPFS). Elastic Storage melds massive ingest rates, flash storage support and automated storage management capabilities, enabling businesses to exploit Big Data.

"IBM Elastic Storage provides a global name view across data center locations," noted Tony Pearson, an IBM Master Inventor and a System Storage senior IT specialist. "It can manage up to a Yotabyte of information, combining Flash, disk and tape resources. It supports OpenStack interfaces, Hadoop and standard POSIX file system conventions," he added in summary.

Elastic Storage promises major savings as well. IBM boasts that the software's policy-driven, analytics-enhanced automated tiering capabilities can slash storage costs by up to 90 percent by moving data to the most appropriate storage device.

IBM's software-defined storage play has roots in Watson, the computer Jeopardy opponent that bested former champions in 2011. At the time, Watson's knowledge store included access to 200 million pages of structured and unstructured data, courtesy of IBM's underlying storage technology.

Today, IBM is priming Elastic Storage for the Big Data era. The company reported that in an IBM Research demo, the software successfully scanned 10 billion files in 43 minutes using on a single cluster. The feat, said IBM, "translates into unequalled performance for clients analyzing massive data repositories to extract business insights."

The solution also leverages server-side flash to boost application performance. It detects flash capacity on a server and automatically enlists it as cache memory. IBM joins companies like EMC (via VMware) and PernixData to deliver high-performance shared storage capabilities via server-side flash, or in the case of software storage startup Maxta, eliminate SANs altogether.

Elastic Storage picks up where traditional storage left off as organizations struggle to manage their Big Data, said Tom Rosamilia, senior vice president of IBM Systems and Technology Group. "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," he said in a statement.

"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," added Rosamilia.

The hardware agnostic platform can run on a cluster of x86 servers or IBM's own Power-based hardware and supports storage systems from multiple vendors. Elastic Software will also be available later this year as an IBM SoftLayer cloud service, said the company.

Pedro Hernandez is a contributing editor at InfoStor. Foll

Photo courtesy of Shutterstock.

This article was originally published on May 12, 2014