TwinStrata joins the market just after other startups like Nasuni Corp. and StorSimple Inc. entered the same general field — providing local storage caches and management features to front cloud data storage repositories. Like those companies, TwinStrata plans to offer what it calls intelligent caching, in which the most active data remains local while static data remains in the cloud and data is pulled back and forth between the two repositories automatically, according to either policy or frequency of access.
However, TwinStrata co-founder and CEO Nicos Vekiarides said the company plans to differentiate its CloudArray software by offering granular caching policies that allow different data to be stored on different types of local hardware on-premise and be cached according to different rules.
TwinStrata will also look to stand apart from competitors by offering users the ability to create, migrate and access applications running on public compute cloud infrastructures. TwinStrata can integrate with compute services like Amazon's Elastic Compute Cloud (EC2) to allow users to run virtual machines(VMs) in the cloud or stand them up in the event of a disaster.
TwinStrata is also looking to differentiate itself by offering features within public clouds and providing a connection between the user's premise and public cloud data storage services. TwinStrata's CloudArray software can look to a file similar to a database transaction log within public cloud data storage repositories and use it to create crash-consistent snapshots on an hourly, daily, weekly and/or monthly basis, storing only delta changes and using thin provisioning in the cloud.
If a user has an account with two public cloud data storage service providers, CloudArray can also create snapshots from one cloud data storage repository and store it in another public cloud. CloudArray can also be used to migrate data between public clouds. Vekiarides said transaction-consistent snapshots, including integration with Microsoft's Volume Shadow Copy Service (VSS), is on the roadmap.
In addition to integration with Amazon's Simple Storage Service (S3) and EC2, TwinStrata has so far struck up partnerships with EMC Atmos, Atmos partner AT&T, Hosted Solutions and Mezeo Software Corp. Vekiarides said provider-specific policies, such as AT&T's option of one-site or two-site replication, can be activated through CloudArray.
One early adopter, Boston-based IT consulting firm NSK Inc., will also become a value-added reseller of CloudArray after using it for the last several months. According Alex Straffin, NSK's senior helpdesk associate and virtual infrastructure manager, the company had begun to outgrow its approximately 5 TB direct-attached storage (DAS) environment, which contained a number of redundant copies of files without a networked storage system in place. Today, NSK has streamlined its cloud data storage to 1 TB between Amazon S3 and its premises. "For every 100 GB of data that's out there, we're only storing 10 GB to 20 GB locally," Straffin said.
Terri McClure, a senior analyst at Milford, Mass.-based Enterprise Strategy Group, said TwinStrata's product is part of the evolution of file virtualization products she and other analysts recently predicted, but "[CloudArray] is much broader than file virtualization, [and] more than just a connector to the cloud. Cloud-to-cloud migration is a differentiator. Mirroring between clouds and block data support are also key advantages for TwinStrata today."
While the latest "gold rush" is on among data storage vendors looking to corner the cloud storage market, it's unclear what percentage of enterprise organizations will embrace the new medium. Many enterprise storage pros cite compliance and data security concerns in addition to concerns about performance of cloud data storage.