IBM Intros Software-Defined Storage Platform

By Jeffrey Burt  |  Posted 2014-05-12

IBM Intros Software-Defined Storage Platform

BOSTON—IBM is using technology leveraged from its Watson cognitive computing initiative and its work in high-performance computing as the foundation for its first significant foray into the software-defined storage arena.

At an event here May 12, IBM executives unveiled "Elastic  Storage," a technology they said will address the needs of businesses and service providers that are being inundated by the skyrocketing growth of data being generated, not only through traditional means, but also newer venues like social media, mobile and video. Elastic Storage is designed to let businesses quickly and efficiently store, access and analyze all the data that is coming in, regardless of the source.

The new software-defined storage (SDS) offering, initially developed for the Watson artificial intelligence project that gained national fame via its victory on the game show "Jeopardy," offers significant performance and scaling advantages over traditional storage techniques, and automated tiering capabilities—stashing data on the most efficient and economical storage hardware, from tape to solid-state disks (SSDs)—that will help drive down storage costs by as much as 90 percent, according to IBM officials, customers and partners.

"Big data has become bigger data," Russell Schneider, principal storage consultant at IBM partner Jeskell, said during a panel discussion at the event here.

Schneider noted that the data businesses need to manage is not just being generated in the data center, but in myriad places in the world. Enterprises need to be able to rapidly bring that data together, regardless of where it's created, and manage and analyze it.

IBM is extending its reach in a nascent but growing and highly competitive part of the storage business. Competitors like EMC—which on May 5 introduced its ViPR 2.0 SDS platform—have been talking about software-defined storage for more than a year, part of a larger trend in the industry toward software-defined data center architectures.

"'Sofware-defined' means a lot more flexibility," Tom Rosamillia, senior vice president of IBM's Systems and Technology Group and integrated supply chain, said during the event. "It means your system can expand very, very quickly. … There are incredible amounts of data that need to be accessed quickly."

IBM already had storage virtualization technology that could scale beyond what competitors could—it can virtualize 250 appliances—Jamie Thomas, general manager of storage and software-defined systems at IBM, told eWEEK. Elastic Storage represents Big Blue's most significant foray into SDS, Thomas said.

The new offering supports next-generation, data-intensive applications and targets data in the file and object formats, she said. It leverages IBM's Global Parallel File System—including its access, management and governing capabilities—and virtualizes physical storage to create pools of shared storage. IBM officials also stressed the solution's openness, including support for the OpenStack cloud orchestration software, Hadoop and POSIX.


IBM Intros Software-Defined Storage Platform

Businesses using Elastic Storage can manage, access and analyze data of any type—structured and unstructured—on any storage hardware and on systems from IBM and its competitors.

"So the new technology can support a range of computational processes, including real-time analytics on virtually any form of data regardless of the underlying storage technologies, platforms and vendors," Charles King, principal analyst with Pund-IT Research, wrote in a post on the firm's blog. "Practically speaking, this means that rather than choosing/moving data to individual systems for transaction processing, analytics and other efforts, Elastic Storage can automatically and simultaneously support multiple workloads on the data residing in storage pools without requiring duplication or transference."

IBM appears to be targeting the new offering at such applications as risk analysis and scientific research, "cutting-edge workloads where the need for raw performance often overrides price considerations," King wrote. "As a result, we expect interest in the new solution will mostly reside, at least initially, among companies and organizations already familiar with IBM's Watson, GPFS and other technologies."

The target customer group could expand, he wrote. IBM initially is offering Elastic Cloud as an on-premises solution, but it will be available as a service through the vendor's SoftLayer cloud later in the year.

Chip maker Cypress Semiconductor has been using Elastic Storage after an experiment with an open-source storage solution didn't work out, according to Alan Malek, IT director at the company. Cypress had recently seen "an explosion of data," and needed to find a platform that was resilient and scalable, and would let engineers quickly access the information they need. The company has gotten 8.5 to 13 times the performance over previous technologies, Malek said.

"We have so much more flexibility," he said during the IBM event. "Our applications do not have storage as a bottleneck anymore."

The Elastic Storage technology gave Watson tremendous scaling capabilities during the "Jeopardy" show. Watson could leverage the Elastic Storage technology to get access to 200 million pages of unstructured and structured data, and about 5 terabytes of information—about 200 million pages of data—were loaded into the system's memory in a few minutes, Thomas said during the event. IBM researchers found that the storage solution can scan 10 billion files on a single cluster in 43 minutes.


Rocket Fuel