41–50 of 53 posts
They haven't had a single SSD failure since. Granted, their search index is read only.
Consider this: the hard drive is the least reliable component of almost any modern PC/laptop. Add to this that it likely contains the most valuable non-commodity asset: your data. Anyone not backing up their system is really asking for trouble... which will happen. Given the ease of use of modern backup systems, and their cost (free only costing a modest amount of your time), everyone should be doing it. OSX and Win7…
> Consider this: the hard drive is the least reliable component of almost any modern PC/laptop. Is it really? Completely anecdotally I've never lost a HDD, but keep going through power supplies… it'd be nice to see some data on component failure rates. I assume some exists, but I haven't really felt strongly enough about it to go look.
At the moment, the reliability problem with SSDs stems entirely from firmware bugs, rather than the underlying flash technology. All the issues you hear about with regards to drives causing blue screens or simply failing to be recognized by the system at all after a while are issues with the firmware on the controller chip - the actual flash chips themselves are pretty dumb and rarely fail catastrophically. This will…
It would be nice to see the SSD manufacturers get a clue about recovering data. Having the device fail catastrophically is nuts; it means that all of the carefully designed recovery schemes in the file system are basically worthless. Yuck.
However, bigger and bigger enterprise vendors are doing rigorous qualification tests on SSDs for white-labeling purposes and demanding that the firmware be bulletproof. Admittedly the enterprise workload is a bit different from a laptop (servers never go to sleep) but they are the ones accumulating the most runtime given 24 hour uptimes.
Consider this: the hard drive is the least reliable component of almost any modern PC/laptop. Add to this that it likely contains the most valuable non-commodity asset: your data. Anyone not backing up their system is really asking for trouble... which will happen. Given the ease of use of modern backup systems, and their cost (free only costing a modest amount of your time), everyone should be doing it. OSX and Win7…
I disagree. In my collection of always-on 30 machines in my basement, over the last three years I have had probably 10 power supplies die, and only two hard drives. The hard drives that died--two 500 gig drives--died in the same month after just two years. I don't keep spare hard drives, but i do keep spare power supplies.
Perhaps your situation is more analogous to servers (most mid-high end servers come with redundant power supplies).
One thing the article didn't address is performance over time. Even with TRIM support, SSDs get slower as they're used. Occasionally, cells die prematurely causing spare capacity to decrease. This won't affect read speeds much but it will hurt writes. These older SSDs will still be faster than hard drives but they won't be as fast as you'd expect them to be. Now for some of my own data. Here's an Intel X25-M G2 after…
Once upon a time, it was best practice to keep the swap file/partition (or log files, etc.) off of the SSD in order to keep the frequent writes from wearing out the SSD too quickly. Do you feel like this is outdated advice?
Consider this: the hard drive is the least reliable component of almost any modern PC/laptop. Add to this that it likely contains the most valuable non-commodity asset: your data. Anyone not backing up their system is really asking for trouble... which will happen. Given the ease of use of modern backup systems, and their cost (free only costing a modest amount of your time), everyone should be doing it. OSX and Win7…
They have improved, but I do not think we are there yet. For example, I do not know of any system that regularly verifies backups. Time Machine, for example, could lose a directory of photos from five years ago without noticing it. If, then, your main disk fails, The photos are, for all practical purposes, gone forever.
Why is the IT industry so cautious about SSD reliability? We have spent decades developing HDD fault-tolerance mechanisms or processes such as RAID and backups. We should trust them .
As SSDs are maturing it looks like there are two distinct markets where hard-disks can't go:
1) In servers, vendors offer cards that tie directly into the PCIe bus, bypassing SATA (
2) For laptops, one can look at the Macbook Air and it's SSD chips. Right now they are connected via SATA, but if that becomes a limiting factor, I see how that would be bypassed as well.
For now HDs and SSDs are very similar, but SSDs (which are essentially flash memory) can and will go where HDs cannot. They are quite different.
I wish people would compare these with IO operations instead of wall time. SSD is a huge win there, and that matters the most for a lot of people.
One thing the article didn't address is performance over time. Even with TRIM support, SSDs get slower as they're used. Occasionally, cells die prematurely causing spare capacity to decrease. This won't affect read speeds much but it will hurt writes. These older SSDs will still be faster than hard drives but they won't be as fast as you'd expect them to be. Now for some of my own data. Here's an Intel X25-M G2 after…
Consider this: the hard drive is the least reliable component of almost any modern PC/laptop. Add to this that it likely contains the most valuable non-commodity asset: your data. Anyone not backing up their system is really asking for trouble... which will happen. Given the ease of use of modern backup systems, and their cost (free only costing a modest amount of your time), everyone should be doing it. OSX and Win7…
> Consider this: the hard drive is the least reliable component of almost any modern PC/laptop. Is it really? Completely anecdotally I've never lost a HDD, but keep going through power supplies… it'd be nice to see some data on component failure rates. I assume some exists, but I haven't really felt strongly enough about it to go look.
That said, on my desktop I run a raid-1 setup. When a drive fails, I immediately replace the whole pair. Just lost the third one in 6 years this Saturday. And there isn't even any infant mortality distorting the stats.