Both are part of IBM’s on-premises Cloud Object Storage and will be generally available Dec. 1.
Object storage manages data as objects, as opposed to files or blocks, and is typically used in the cloud.
“In the cloud, object storage is the standard, de facto storage architecture for persistent data,” said Rob McCammon, offering leader at IBM. “On-premises, most data still sits in block storage, or for longer term, file storage.”
However, enterprises are increasingly moving to cloud-based (both public and private) SDS, which typically lowers costs and is easier and faster to provision compared to legacy storage architectures. In a study published earlier this month, IDC forecasts worldwide SDS revenues will reach nearly $16.2 billion by 2021, experiencing a compound annual growth rate of 13.5 percent between 2017 and 2021.
As companies move to the cloud, either using entirely cloud-native storage or a hybrid cloud strategy, “part of that will require them to change their architectures to utilize object storage,” McCammon said.
IBM calls the first new SDS feature “compliance-enabled vaults.”
Certain regulations like SEC 17a-4(f) require companies to store some data immutably, preventing it from being deleted or modified for a specified retention period. The new software capability allows companies to meet these requirements by preserving electronic records in a non-rewritable and non-erasable format.
“While we’ve initially been focused on the needs of our clients in the financial services market — and have started working with one of the top 5 U.S. banks that plans to use the compliance-enabled vault capabilities — we also see strong demand in other markets,” McCammon said. “These include health care and government, when you are talking about things that might be needed for evidence like bodycam videos.”
The second new software feature allows companies to deploy entry-level cloud object storage systems with the ability to scale to larger configurations.
Companies can now deploy these systems at a capacity as small as 72 terabytes (TB). IBM used to require a minimum of around 200 TB capacity.
This concentrated dispersal mode capability also enables smaller footprint systems. “It requires 75 percent fewer servers to initially start your cluster, and gives you the opportunity to start smaller at a much lower cost,” McCammon said.
This allows companies that are new to object storage to start small, and grow to petabyte scale and larger.