For many people working in IT, it seems unthinkable that networking is a relatively recent phenomenon. Prior to this, information was exchanged primarily on floppy disks—limited, unreliable and with zero security. Unsurprising, networking quickly asserted its superiority over the manual alternative.
But while networks quickly enabled basic file, data and printer sharing, the concept of really networking data access didn’t arise until the Storage Area Networks (SANs) of the late 90s and the initial Network Attached Storage (NAS) alternatives that followed. The problem here for the SMB level companies is that these solutions were simply far too expensive, explains Chris Davies, General Manager D-Link UK.
Now, however, NAS or Network Attached Storage has become affordable even for small businesses. So, rather than simply storing data on individual PCs, offline tape backup (complex and impractical in many ways) or?worst of all?on individual USB memory sticks (horrendous security issues), it makes a huge amount of sense for an SMB to add network attached storage and gain reliable, secure data storage that is readily accessible. However, it is important to invest in storage technology with the right level of features and functionality, which we will now cover here.
So why invest in networked storage? Well, there are some fundamental advantages?network storage consolidates data, both literally and in terms of management costs. It provides a secure, central data repository that is available locally or remotely, 24×7. Equally important are the considerations of what might happen if that facility is not available. The starting point here is data loss. So many companies now rely heavily?if not entirely?on their network and data, that any data unavailability quickly becomes very costly.
Complete data loss can be a genuine disaster. Worse still, this can be a result not simply of accidental deletion or failed components, but through security breaches and wilful data destruction. A joint study by PGP and Symantec on data loss revealed that data breaches cost UK companies an average of £47 for every single record lost. What this amounts to is that the average cost to a company suffering a data breach is £1.4m. This was based on a study of UK companies?these are not figures made up out of the blue.
The size of the losses examined ranged from 2500 records to in excess of 125,000 and costs incurred ranged from £84,000 to £3.8m. Breaches by third parties were more expensive than in-house losses. On average they came to £59 rather than £42 for in-house, per record. This is a difficult issue for companies to deal with, since their supply chain might include hundreds of partner and outsource companies. The cost from lost business in the wake of a data disaster is 36% of that total data loss with financial services companies naturally being particularly at risk with far higher costs. The estimate is that customer churn rates go up by an average of 2.5% after a data loss, but the worst example in the UK reportedly saw churn rates increase by 7%.
Security is therefore a key requirement of data storage, especially when that data is being shared by a wide mix of different user types and particularly when it is being accessed remotely via the Internet. With FTP still being a widely used protocol for file transfer, it’s important to understand that basic FTP is inherently insecure, so a Secure FTP (SFTP) option is a vital feature. Equally for any server functions, supporting SSL for secure, HTTPS connections provides a completely secure back-end.
Data encryption not only secures the data, an obvious benefit, but may well become mandatory in the future. For the users themselves, it is important to be able to provide different access rights to different user types and levels?for example, read only or read/write.
Data security in the sense of availability and redundancy is also critical, so, look for RAID support. RAID comes in various formats, such as 0, 1 and 5, so the more modes supported, the more flexible the solution to fit the specific business needs. For example, RAID 0 combines the drives in a ‘striped’ configuration, and is the choice offering the highest performance. RAID 1 causes the drives to mirror each other, providing maximum protection. If one drive fails while configured as RAID 1, the unaffected drive continues to function as a single drive until the failed drive is replaced. The new drive will then be remirrored, allowing the storage device to return to its previously secured state. RAID 5 allocates data across three or more drives and combines storage efficiency with reliable file protection.
This kind of flexibility is key to all aspects of networked storage, not least scalability. With new wave applications such as video and IP surveillance generating very serious amounts of data, it is vital not to underestimate how much storage space is required, not just now, but in 12 and 24 months time. So, the ability to add extra drives to your network storage is key, rather than having to reinvest in brand new devices every time you want to increase your storage capacity.
IP surveillance, for example, is an increasingly important application for SMBs, not simply as a cost-effective and easily managed alternative to traditional CCTV, but?because of its flexibility?the basic components, such as high-resolution Web cameras, can be used in many different kinds of applications such as Internet conferencing and training. But regardless of the exact application, the important point to bear in mind is that when using devices such as 10-Megapixel cameras, the amount of data being generated is huge. So the storage capacity has to be able to match that data generation of the applications themselves.
Here is also where specialist storage such as NVRs (Network Video Recorders) have a significant role to play. For all kinds of video applications, such as those mentioned above, the NVR removes the need for PCs to store data, so that events, training and other video recordings can be centrally stored and made available online at any time, with the video delivery completely under the control of the user. Some general network storage is also available with video streaming capabilities, so this is an important point to note for some, as part of the evaluation process.
One final point to make is that, regardless of the feature set, the network storage should be easy to setup, configure and manage. So, look for plug ‘n play capabilities, auto-assignment of IP addresses for instant connection to the network and a simple management GUI to help reduce management and administration costs. Other features such as support for Active Directory may also be important differentiators.
Overall, the argument in favour of networked storage for any SMB is overwhelming. Providing secure, centralised data, accessible from anywhere, manageable from anywhere, is an obvious advantage. Add in key elements such as scalability, data redundancy, support for new wave video applications and the future-proofing is complete.