IDC on Wednesday said the worldwide market for file-based and object-based storage is gaining momentum, and that cloud vendors building their own object-based storage systems is driving this market to grow faster than the commercial storage business.
IDC forecast that worldwide revenue for file-based and object-based storage will reach $38 billion by 2017, a huge jump from the market's estimated $23-billion-plus revenue in 2013.
Object-based storage is a way of storing data as flexible-sized objects rather than the fixed block sizes used for storing data in traditional SAN arrays, making it more suitable for handling unstructured data or data in a cloud.
[Related: 1Q Storage: Dell Sales Boom, HP Sales Bust]
Object storage, as defined by San Antonio-based hybrid cloud provider Rackspace, doesn’t provide access to raw blocks of data or file-based access to data. Instead, object storage provides access to whole objects, or blobs of data, generally via a system-specific API.
"Object storage excels at storing content that can grow without bound. Perfect use cases include backups, archiving, and static web content like images and scripts. One of the main advantages of object storage systems is their ability to reliably store a large amount of data at relatively low cost," Rackspace wrote.
Object-based storage has become a huge part of the storage industry, thanks to the growth of companies that in the past were not counted as storage vendors, said Ashish Nadkarni, research director for storage systems at IDC.
While traditional disk-based scale-up storage sales is starting to contract, scale-up and scale-out file-based and object-based storage sales are forecast to grow at a compound annual growth rate of 24.5 percent from 2012 to 2017, Nadkarni said.
That growth stems in large part to the increased use of the cloud for object-based storage, he said.
"A lot of cloud providers who provide object-based storage like Amazon and Google are not buying storage equipment from companies like EMC or NetApp," he said. "They buy their own disks from companies like Seagate and get their servers from white-box vendors. And they have their own file systems."
"It sounds like a good business to be in," said John Woodall, vice president of engineering at Integrated Archive Systems (IAS), a Palo Alto, Calif.-based solution provider that works with object-based storage via partnerships with Hitachi Data Systems (HDS) Content Platform, NetApp's Storage Grid, Dell's partnership with Caringo and SGI's partnership with Scality.
NEXT: Making Sense Of Object-Based Storage Growth
"It's still in some ways an early market," Woodall said. "There's lots of startups."
Traditional RAID-based scale-up storage systems are good for storage capacities of hundreds of gigabytes, or maybe a few petabytes, Woodall said. However, as customers' unstructured data files or large structured data files continue to grow, the importance of object-based storage grows.
"For example, in the medical field, you can have an object store around records, including information on who can access the data, when it's accessed, what the insurance is," he said. "Your doctor might see the X-rays and test results in the record, but the claims processor may only see the billing data. That data can be tuned for different retention periods or data protection solutions. The records also need metadata to classify the different parts, to make copies, or to replicate it based on policies. That's not traditional storage."
Several vendors offer scale-out file-based and object-based storage technology, Nadkarni said.
On the file-based storage side there are such vendors as EMC Isilon, NetApp Cluster Mode, HP Ibrix and Red Hat's Gluster technology. Object-based storage providers include Data Direct Networks, Scality, Cleversafe, Amplidata and EMC Atmos, he said.
"But the biggest chunk of object-based storage is the self-build companies including Amazon and Google, which accounts for 90 [percent] to 95 percent of all object-based storage," he said. "When Amazon and Google build their cloud, they build with petabytes and petabytes in mind."
PUBLISHED AUG. 8, 2013