For a while now I’ve been considering alternative methods for storing my photo library which now has hundreds of thousands of photographs and is growing at what seems to be an exponential rate.
Currently I purchase external harddrives of the same brand and same capacity in pairs and mirror the information across them incase one fails. So for example: I own two 1.5 TB external harddrives, and two 3 TB external harddrives. The two 1.5 TB drives are nearly full with – at the time of this posting – 15 GB of free space remaining on each, while the two 3 TB harddrives are empty — they will be the next drives I move on to after my 1.5 TB’s are full.
Here’s a quick version of my workflow:
- Shoot photos, copy from CF card into a folder on laptop’s SSD
- Repeat this process for the entire month, backing up each shoot to a folder on a portable external HDD.
- At the end of the month:
- Copy folder on SSD containing the photos from that month and paste it onto one of the 1.5 TB drives. Disconnect that 1.5 TB drive, plug in the other, and repeat.
The folder of images averages around 50GB of space and takes about 20 minutes to transfer via USB 2.0 (because both 1.5TB HDD’s use USB 2.0). But now that they’re nearly full I’ll be moving on to my 3 TB HDD’s which use USB 3.0, so the transfer time for a folder of the same size – 50GB – is reduced to about 7 minutes.
So I’ve thought about this process and how it could be improved – and there are ways it could be – but let me explain why I like it..
I’ve been shooting since 2007 (albeit not full time as I am now), but even so, it’s taken me 5 years to fill up a 1.5 TB drive. If I buy something like a Drobo with 10 or even 5 TB of space I can’t imagine how long it would take me to fill up those drives, and the biggest reason I’m hesitant to do that is because the rate at which computing technology changes. For instance, the Drobo 5D is one of the latest models available from Drobo. It’s pretty awesome because it has thunderbolt capabilities, which would be useful for my purposes since my 2012 15″ Macbook Pro Retina has thunderbolt built in. As I’ve established, it took me 5 years to fill 1.5 TB… so we’re talking about a long, long time to fill 10TB of space, and computing technology would change at a far faster speed than the rate at which I could hope to fill 10 TB’s of space. So while although the Drobo 5D currently uses thunderbolt, which right now is really fast, who’s to say in 6 years – when ‘theoretically’ I may have 3TB of data filled on the Drobo – that it will still be the fastest data transfer?
Another downside to the Drobo is that although the data is mirrored within the RAID it’s still physically all in one place. So if ever there was a fire at my home, or if the Drobo was stolen, or damaged, all of the content would be gone. Then, there’s the expense of the Drobo. Right now the Drobo 5D is $800, then you have to buy the drives to go into the Drobo. So if I wanted to stick with my example – 10 TB of space – I’d be looking at spending roughly $1500-$1700 for everything.
The reason I prefer the data back-up method I’m using now (mirroring independent harddrives) is because it solves a lot of the issues I foresee with the Drobo. For example: the costs. To buy 10 TB of space for the Drobo – right now – I’d be spending quite a bit of money, but in the future, say, 4 years from now, 10 TB of space will probably be in the same range that you’d pay for a 4 or 5 TB external harddrive. So it’s easier on my wallet, which is always good. Second, if ever there was a fire or theft or whatever else, the data is contained across two separate, physical hard drives, which means I can keep one off site. This is probably the biggest advantage to the system I have in place now; the ability to move and keep one of the mirrored drives off site. Thirdly, the issue of speed: The Drobo 5D uses thunderbolt – which like I said is currently the fastest available data transfer – but when you buy the Drobo 5D unit ($800), you’re more or less stuck with it for the long haul, whereas with the external harddrives that I’m using now, updates and improvements reflect the industry standard for the computing environment meaning USB 2.0 was dated, so USB 3.0 was developed, and eventually that will be replaced, too. Going forward, it just seems that the speed, cost, and ability to mobilize and store content off site, having separate harddrives seems to be the best solution.
One other last announcement I wanted to make in this blog:
I’ve been working on a new way to keep in-touch with my readers which I think is going to be really awesome, although its been a headache to set up. I’m nearly done and I think I’ll be rolling it out sometime late this month or in early January. I’m really excited about this and I think it’s going to be a really useful way to provide all you with up-t0-date and somewhat exclusive content, so stay tuned.
If you have any thoughts on this blog or about the way I store my data, or any tips or advice about how you’d do it better, let me know in the comments section.