Desktop NAS: What it is and what it’s good for

Desktop NAS provides bulk storage with basic RAID protection for small and branch offices, but what are its limitations, and can it be used for backup and primary data?

Desktop NAS devices offer cheap and easily configurable storage capacity for small-office and branch-office use cases. Usually these devices come with basic RAID data protection and the ability to link to network directories to set security provision. But, what are the limitations of desktop NAS, and how useful is it as a backup target or for primary storage?

In this interview, Bureau Chief Antony Adshead speaks with Martin Taylor, converged network manager at the Royal Horticultural Society, about the key characteristics of desktop NAS and the use cases it is best suited to.

You can read the transcript below or listen to the podcast on desktop NAS systems. What is desktop NAS and what features and levels of performance are available in this kind of hardware?

Taylor: To start from the base level, we’re talking about a small disk array that’s capable of RAID striping, generally only two levels, which will be RAID 10 or RAID 5 depending on the resilience that you want, with a native operating system installed on the box that acts as a broker between your clients and the file serving. Quite often, these include things like Active Directory connectors so you can interrogate your AD for security information.

The first network-attached storage device I ever worked with was a SnapServer that had something like a 200 file share limit on it, so obviously things have come on a huge amount since then. NAS is basically transparent to users these days. There’s very little overhead for talking to your Windows network, and it’s a cheap and easy way to store files that require file service access. Not in the same way as direct-attached storage. Obviously, if you buy something with iSCSI connectivity you can begin to host applications and databases but for the purposes of this discussion we’re talking about file serving.

If we’re talking about levels of performance in desktop NAS systems, they seem to be either four- or five-drive-capable with Level 10 or Level 5 RAID. This seems to be common for most of the entry level devices such as the [Buffalo] TeraStation.

It’s very easy to configure these devices for users. There are simply drop-down menus and you manage them via a Web interface. They tend to have a broker operating system on them that talks to Windows. This can be Linux-based [although] there are flavours of operating system that come direct from Microsoft that are attached to these devices. Again, the important thing here is linking into your Active Directory if you’ve got that requirement.

If [you’ve got] a very small office or are a home user, you can put this device on your network and point your files at it and off you go; it doesn’t require any security configuration. But if [you’re in] a small-office enterprise environment where you need to define levels of security access to files, then it’s dead easy because there is nothing native on the box and you can do it all through Active Directory. You set file and folder permissions through the interface on the desktop NAS device that are dragged from your AD. What can you use desktop NAS for, and what about its suitability for use for backup or primary storage?

Taylor: In the small-enterprise environment that I’m in, desktop NAS is a very cheap and easy way of gaining some space for file archival or email archival purposes: stuff you need quick access to [if you’re at] a small remote site. Also, image storage is a big thing with us with our large image databases, and we find they’re a very cheap way of storing large amounts of images; they are available to users all the time. And there’s redundancy in the devices so we don’t have to worry about failures, and we get alerts if there is a problem.

So, basically, they’re great as a bin area for all files that you don’t want on levels of expensive storage. It’s perfectly possible to drop files down to them off of the SAN as well, when we find it’s no longer economic to keep them there for access reasons, etc.

But mainly the use for us, at small satellite offices, they’re very cheap file storage, even cheaper than Windows storage. When you look at the price of storage on desktop NAS you’re looking at something comparable to a single server drive, so with the amount of disk space you get and the redundancy, economically desktop NAS definitely makes sense in the [small-enterprise] environment but it must be used correctly. I certainly wouldn’t consider it for live, online storage of files that required a lot of access; that would be more server-based.

The other use we find for desktop NAS in our environment is that it’s a very cheap way of hosting backup files. Currently, we’re using [Symantec] Backup Exec … to back up all our stuff and find that desktop NAS is good as a backup to disk device. This is for non-critical backups, incremental, etc.

Obviously, we hive off our full backups off to tape but [the desktop NAS is] great as transition area, as compared with the cost of storage such as tape the price is very favourable and because it’s online all the time, the restore times are quick. Because there aren’t any problems getting people to deal with tape loading [at] remote sites, etc., then it’s available instantaneously … if we have to restore something for users. So as a backup staging area I definitely think they have a good use, and that’s the main purpose we use them for other than file storage.

Read more on SAN, NAS, solid state, RAID