Network Storage Comes of Age
How the network storage appliance is changing to meet the demands of the growing business
Once seen purely as a small business product, the network storage appliance has, in the last few years grown both in capacity and, more importantly, functionality. Features once only found on high-end storage solutions are being incorporated, often at no extra cost, enabling vendors to offer much more rounded solutions capable of meeting the storage needs of companies large and small.
First there was NAS
Most small businesses will be familiar with NAS (Network Attached Storage) devices, designed to replace general purpose file servers with something more affordable and easier to manage. These “black box” solutions deliver access to storage direct to client computers in the form of network file shares - just like a Windows, Mac or Linux file server but with a much simpler Web-based interface for setup and management.
Available in multiple guises from simple consumer NAS boxes to more scalable rack-mount appliances, RAID (Redundant Array of Independent Disks) has become commonplace making NAS suitable for use on small business networks both for general file sharing and to handle desktop backup. However, there are limits when it comes to what NAS can do, especially in areas of scalability and performance. Limits which may not bother desktop users but can be a real issue when it comes to servers.
Then came iSCSI
As NAS appliances can never replace Windows and other servers altogether, they are alive and well, and to be found hosting business-critical applications across the globe. Moreover, on the back of virtualisation, there has been an explosion in demand for fast, flexible and reliable storage to go with them.
The old way of meeting this demand would be to deploy a separate Fibre Channel Storage Area Network (SAN). The newer, simpler and, incidentally, more affordable way is to provide the same block-level access to shared storage as with a Fibre Channel SAN, but using the iSCSI protocol over an ordinary Ethernet network infrastructure.
In less than a decade, iSCSI has become the protocol of choice for small to medium-sized SAN deployment and it’s easy to see why. Just as with Fibre Channel, iSCSI storage can be provisioned and managed centrally, yet presented as virtual volumes that appear just like directly attached disk drives as far as client servers are concerned.
Unlike local storage, however, iSCSI volumes can be created in seconds; re-assigned to different servers at will; grown in capacity; shared on server clusters; replicated and copied for extra security. And all using standard Ethernet switches and cabling for connectivity.
Two become one
Given such benefits it is little wonder that network storage vendors have latched onto iSCSI and, in the past couple of years, have begun to add it as an option on business-class storage solutions.
Pure NAS boxes are still available, of course, as are appliances that only handle iSCSI. In addition, however, vendors are increasingly offering unified products, like the D-Link ShareCenter Pro 1550, capable of simultaneous NAS and iSCSI SAN sharing.
Still managed via a simple Web interface - just like a NAS box - unified storage appliances enable customers to provision both network file shares and the virtual volumes needed to support email, web and other application servers using just the one device.
High availability and more
Of necessity, along with the benefits of iSCSI, come additional responsibilities and requirements. When used to support business-critical applications, for example, buyers of unified storage appliances will be looking for the kind of high-availability features traditionally associated with more expensive enterprise class solutions. These are features that enable mission critical systems to survive common hardware failures and recover quickly when disasters occur.
Basic RAID is not enough here. Customers want products that can withstand the failure of two disks at a time (RAID 6). Likewise, they want to know that the storage appliance can carry on working even when a power supply fails. And that means redundant power - not always an option, but something the ShareCenter Pro 1550 delivers as standard.
The affordable D-Link appliance also features other high availability features normally associated with products costing a great deal more. Features such as the ability to automatically replicate storage content across the network to protect against possible disaster, and take instant snapshots which can be used both for backups and to recover quickly when system failures occur.
Data encryption is, similarly, available on the ShareCenter Pro 1550 and, as with all of these extra features, is included in the basic price rather than requiring an additional licence.
Of course, there will always be a fine balance between price and the feature set of storage appliances marketed at small to medium-sized businesses. That said, as the technology of iSCSI matures and becomes even more affordable, you can confidently expect other high end features to trickle down and put in an appearance on an appliance near you very soon.