the little d disasters

SSDS: A blessing, a curse and a cautionary approach

Just built another systems up, with a Samsung 840 240GB SSD. While I’m used to spending the next few hours wandering by the bench, clicking restarts, get updates, etc (some of you know the drill well), it was strikingly different last night.

I pieced a small form factor Dell Optiplex together (and saw issues with cable routing and screw heads and fan blades – another post worth making) and fired it up with the install disk inserted.

About 1.5 hours alter, I had gone from Win7/SP1 to all updates loaded (more than 103), and application software loaded (Office, Reader, Flash, and security). Usually I get most of the way there and then it’s a few more overnight hours and a few more reboots first thing in the morning.

I’m getting jealous of these systems, as they go from off to Desktop in about 15 seconds…

Oh, an high speed cable connection (about 39M download) didn’t hurt either, but I’ve used it for builds before, too.

As I was being dazzled at the rapid completion of progress bars, I reflected on a comment from a shop owner here I occasionally drop in on regarding his experience as an early adopter.

Kirk mentioned he had gotten and SSD about a year ago. It was great, but one day it was dead…as in really, really dead. Think about it: While disk errors send shivers up our spines, unless we hear that horrible grinding, and/or screeching noise, we techs have a degree of comfort that the data is not really gone, just accessible as a non-boot drive, and all is mostly right in the world.

An SSD? It’s common failure mode is DEAD! no connecting via a USB adapter (which is ever present with me, like a credit card, when I leave the house), as it’s DEAD!

Blessing: FAST!

Curse: Dead usually really means dead.

Cautionary approach: Back up, back up, and back up. And did I mention backup? Expensive? Not near as much as a data recovery specialist…and I haven’t even done my homework to see if that’s a function out there yet (I suspect it is, and I also suspect it’s costly).

How to resolve the risk? RAID 1. Bite the cost bullet and get two, if you’re going to get one…and than have a “conventional” drive the same size you can clone to, or a partition on a larger drive you can image to….Couple that with an offsite cloud service to ensure a redundant, real time data set is stored for a rainy day SSD drive failure.

Those who know this know I don’t need to say anything else on the topic. In many ways, it’s current common sense, even for conventional, mechanical drives, but it’s one now where a safety net for data recovery from the local failed drive isn’t really there anymore.

UPDATE 3/8/13: As I stated about, the end of life of an SSD drive is way closer than we’re used to when it begins to let you know it’s about to fail, as discussed in this article at MakeUseOf: Can Data Be Recovered From A Failed SSD?