The Next Evolution of Data-Aware Storage
I recently had the opportunity to be briefed on DataGravity’s freshly released Version 2.2 update. It comes with a number of really interesting features to extend the already impressive capabilities of the Discovery Series arrays. Of note in this release: Hyper-V support, file fingerprinting, user activity alerting, some new tags included by default, and various other improvements. Of these additions, Hyper-V support likely has the most impact on the market, but the other additions will make existing users quite happy. Thankfully, my guide, David, took the time to show me all of these neat upgrades!
With the 2.2 release, the Discovery Series arrays now support all of the same features and capabilities on Hyper-V that they previously supported with VMware vSphere. This also includes support for ODX (Microsoft’s offloaded data transfer technology, similar to a subset of VAAI). While customers could previously use the array to support home directories and file shares integrated with Active Directory, they were limited to analyzing VMware virtual machines on NFS datastores introspectively. This new release expands that capability to iSCSI datastores, and be extension, Hyper-V.
Adding support for Hyper-V unlocks a whole new segment of the addressable market that will also benefit from the advanced security, compliance, and efficiency features of these arrays.
This is perhaps the most intriguing feature of this release. Upon ingestion, a SHA-1 hash is computed for a file. By assigning this hash value to the file to identify it as a unique piece of data within the system, the file can be tracked, even if surface-level data about the file changes.
For example, if a Word document called Important Quarterly Financials.xlsx is written to the array, a malicious user could later modify the filename to be logo.jpg and modify some of the operating system metadata that defines the file. On a traditional storage array, this data would be lost. In the event that this malicious activity happened before a backup operation took place, the data would indeed be lost forever.
Thanks to the fingerprinting feature, an administrator can now search the contents of the array for any files with the same fingerprint as the original document. Although the human-readable qualities that identify the example document have changed, the fingerprint would be the same, and the important quarterly financials document would be saved.
User Activity Alerting
Another useful feature in this new release is the capability to set up thresholds and filters for certain user activity that will trigger alerts. A threshold could specify, for example, that if a user changes over 200 files in an hour (read, write, delete, update), it should send an alert. A user changing 200 files in this timeframe is not typical behavior, and could indicate something more sinister like the introduction of the CryptoLocker virus into the environment. Knowing about this sort of activity as soon as possible can be the difference between a small headache and massive data loss.
Another possible use for this might be an alert that triggers when a user reads a specified number of files in a given time frame. Excessive reads could indicate that the intern who was just let go is copying massive amounts of data off the company share before exiting the building. This alert, sent in a timely fashion, could help catch the thief red-handed.
You can learn more about this awesome new release on DataGravity’s corporate blog in the official post about the release. To learn more about the Discovery Series arrays in general, be sure to visit DataGravity over at http://datagravity.com/.