How To Tell If Storage Is Primary Or Data-Aware

Ironically, in conversations about the four pillars of data-aware storage, storage is the element most likely to be ignored or glossed over. This is largely due to the fact that state-of-the-art primary storage capabilities are a necessity for enterprises.

A few common factors define an attractive primary storage array, which DataGravity embodies:

Architecture: The Discovery Series is designed for five nines availability, based on a fully redundant storage platform with no single point of failure, a non-volatile write cache and high availability failover across controllers. Dual-parity redundant array of independent disks (RAID) protection and end-to-end data integrity checking is also included. The system fully monitors and diagnoses itself by writing out system logs, surfacing issues through the user interface, sending proactive system alerts to administrators and forwarding system status information directly to the customer service team. Simply, DataGravity supports mission-critical enterprise deployment.

Performance: The Discovery Series is a native 10 gigabit Ethernet (GbE) device that works just as well in 1GbE environments, supporting high performance access by block and file protocols in VMware, Microsoft and Linux environments. In our hybrid design, more than a terabyte of solid-state cache serve up hot data and metadata for fast input/output (I/O) performance and processing, writes are coalesced into full stripes (and compressed inline) before being written, and a virtualized pool of spinning disks work in parallel to efficiently read and write data across multiple applications and multi-user access workloads.

Cost Efficiency: DataGravity balances the use of solid state and hard disk drives to achieve high performance for working sets of data, while benefiting from the cost economics of denser, lower-cost media. We also exploit high volume server economics by building our system with commoditized, standard and de facto standard parts and system platforms. Finally, by aggressively investing in storage optimization software techniques like compression, deduplication, thin provisioning and free space consolidation, we ensure that the Discovery Series makes the most efficient use of available storage resources.

To an IT or storage administrator, these qualities are a remedial refresher of the baseline features in mid-range storage arrays (even if your incumbent vendor may be lagging behind). Yet, in many ways, these innovations that many take for granted do not address the challenges of the rising data deluge. For this reason, DataGravity is changing the way that people think about the utility of storage infrastructure by creating new standards that go above and beyond the expectations of traditional primary storage.

What features must be added to primary storage in order to consider it data-aware?

Greater Visibility: A key focus area for DataGravity is to enable file analytics within the array, creating insight into how storage resources are actually being used. Who is consuming storage? Is there data that we will never, ever use?  How many files are we storing? What kind of files are we storing? What files can we delete? Without this information, administrators are pained to resolve how to curb storage growth within their environment.

Virtualization Awareness: By establishing virtual machines (VMs) as a key data storage object that the storage array supports, we can then understand, manage, monitor, and protect them alongside data volumes and file systems. DataGravity natively manages hosted VMs, and the platform enables their efficient storage by compressing them inline and deduplicating them on the fly. Additionally, the Discovery Series is aware of each VM’s file system content, providing all of the file-level smarts of our solution to files within VMs – enabling the platform to recover VMs and files within running VMs instantly.

We’ve only scratched the surface as to how data-aware storage will change industry expectations about what storage infrastructure should be capable of. In my next post, I will share how DataGravity is helping organizations with data recovery strategies.

If you like to read ahead, check out our detailed architecture white paper that reviews how we reinvented the primary storage architecture to help storage become fully data-aware.

  Like This

Tim Sherbak

Tim Sherbak leads product marketing at DataGravity, aligned to his passion for helping customers adopt new technologies and solutions to deliver breakthrough results. Tim has held sales and marketing leadership roles in some of the most pioneering companies in technology.