IBM has announced Cloud Object Storage, a service it said lets organizations store large volumes of unstructured data on any combination of their own servers and the public cloud, more cheaply than Amazon Web Services’ S3 service.
The service is built on technology dubbed SecureSlice, which combines encryption, erasure coding and geographic dispersal of data. Erasure coding is a form of data protection. IBM acquired the SecureSlice technology when it bought Cleversafe last year for $1.3 billion. The technology automatically encrypts each segment of data before it’s erasure-coded and dispersed, and the data can only be reassembled where it was originally received.
Under one option, Object Cloud Storage can send treated data to at least three geographically separate IBM cloud regions, while under another it holds data in multiple data centers within the same region.
“It’s technically possible that someone could run apps on Amazon Web Services and store data in the IBM cloud, but I’m thinking most of our business will come from enterprise customers that are using one of our many cloud services,” said Russ Kennedy, IBM’s VP of product strategy, in an interview. Customers may want to move data between the cloud and on-premises servers as their business needs or their perceptions of cloud security change, he said.
IBM claims the new service proved nearly 25 percent less expensive than AWS’s S3 storage service in two internal benchmarks. But it supports the S3 interface, as well as OpenStack Swift.
IBM Cloud Object Storage is available now for enterprise clients using IBM Cloud data centers in the U.S.and Europe and is slated for availability in the Asia Pacific region in December. Availability via digital channels, with credit card support, is set to begin in the U.S. starting in December and in Europe soon thereafter.
Updated 10/13/16, 8:58 a.m. PT, to reflect newly received availability information.