Performance Testing non Windows based NAS devices

I decided to do some very basic performance testing of non Windows based NAS devices that I’ve got here at in my lab / production network.  The aim of this is not to come up with a winner or loser, but to understand what levels of performance I might expect from these NAS devices. I’ll break this testing up into two categories, and tonight’s blog post will feature the non Windows based NAS devices I have.  A future blog post will focus on Windows based NAS devices as well as Windows based Servers.

Currently I’ve got the following devices in my environment.

• Western Digital MyBookLive 2TB – this has a single 2TB drive with no redundancy at all.  Very basic, and just what you can get on a real budget.  There’s a lot more features to this device than just an SMB target, but I’ll go into those in a full review of it later.
• QNAP TS 410 – this has 4 x 2TB drives in a RAID 5 configuration.  It can do a whole lot more than just be an SMB target, but for now that’s all I need from it.
• Buffalo Link Station – this has 3 x 1.5TB drives in a RAID 5 configuration.  I bought this some time back as it was on special from Harris Technology, and I wanted to see what you could do with a 4 bay unit but only have 3 drives spare to put into it for the moment.  This is one of the first tests I’ve done.

Now I’m going to start of by saying that these tests are NOT equal.  Aside from the source machine being my desktop PC, and using ShadowProtect to backup the C: drive, there are many other variations to the configuration.  The intention on this is NOT to give categorical winner/loser, but to get a feel for performance.  I configured an SMB share on each of the devices and then used ShadowProtect to backup my desktop to it.  My desktop has around 121GB data on it’s C: drive, so it’s a good candidate for an average desktop.

Ok – so the results of the test are as follows

Device	Time	Speed
WD MyBookLive	45m 1s	45.68MB/s
QNAP TS410	55m 19s	37.36MB/s
Buffalo Link Station	48m 48s	45.26MB/s

 

What does this tell me?  Well in all likelihood if you want the fastest backup, then a single drive is probably the way to go.  Any form of RAID seems to lessen the performance a little, but interestingly the Buffalo did well with a 3 drive RAID5 array – that surprised me.

So – there you have it – some idea of how fast you can expect from some of these devices when used as a ShadowProtect target.  There’s quite a few other tests that I plan to do with this such as using them as FTP targets for ImageManager, but that will come later.

If any vendor out there wants to loan me a device for testing, I’m more than happy to put it through it’s paces