SSD Write Endurance Considered Sufficient

This is a pretty poor article. Consumers haven't bought SLC in years and MLC is much less reliable (although the actual reliability of flash is still higher than the rating). It doesn't take write amplification or any kind of real workload into account. And it may contribute to the alarmism that it claims to dispel; at worst your SSD will last 172 days? Yikes!

People go completely irrational about avoiding writes to SSD's, I've seen researchers at an EE dept spending evenings tuning their Windows installations and going as far as stuffing frequently used stuff onto a RAM disk.<p>I guess the early off brand SSD's randomly eating your data didn't help the popular perception (even though that wasn't because of flash wear). But still...

I remember when SSD's first started hitting the market years ago, everyone's utmost concern was about wear leveling.<p>As it turns out, if you ship a drive with buggy firmware and it refuses to be recognized after the 50th cold boot, write leveling ends up not being so important.

tl;dr<p>I'm guilty of the SSD fear. My first Intel (320) drive was meticulously maintained in fear of shortened life span and performance losses. I went so far as to question every file copied to the drive vs. mounting an external HDD. The thing bricked itself due to a firmware bug in less than 12 months.<p>Using a Samsung drive now. I don't think about read/writes anymore. I just use the darn thing.

c't has a lang duration test where it became obvious many SSD's were dying long before they should according to the theoretical numbers. Unfortunately it doesn't seem to be available for free: <a href="https://www.heise.de/artikel-archiv/ct/2012/03/066_SSD-Zerstoerung" rel="nofollow">https://www.heise.de/artikel-archiv/ct/2012/03/066_SSD-Zerst...</a><p>This technology is too dependent on software to make these kind of models even remotely approximate reality.

It looks like he's only considering ideal writes with no amplification. This can make a big difference: <a href="http://lwn.net/Articles/428584/" rel="nofollow">http://lwn.net/Articles/428584/</a>

I've always wondered what these theoretical numbers, which has been quoted since long before consumers could ever afford an SSD, should tell you... They should not, in any way, make you trust your SSD. They are beyond useless. Also, this article assumes no write amplification and assumes perfect wear level algorithms. He/She even throws in usb-thumb drives in the mix, so, how do you know if your thumb drive has a perfect (TM) wear level algorithm? What about that flash card? Would you recommend storing swap and log files on that too? Sigh.<p>Compared to this article the advice of not ever storing swap/log files on a flash drive would be sound advice.<p>But yes, you are <i>probably</i> not going to hit the limits of your consumer drive SSD. But whether you are going to or not you will not, under any circumstances, be able to find out by making some numbers up. If anyone knows it is the manufacturer, and you can bet you won't get access to that data.

SMART says the years-old, low-capacity MLC SSD in this computer is 5% worn out. SSDs in RAID arrays at work are getting a much heavier workout and handling it fine.<p>You can use smartctl or the equivalent to check the media wearout indicator to see how much the actual workload has worn out a drive. If your computer isn't constantly writing, you're probably fine--that is, your writes will last until you'd be upgrading anyway.<p>But, keep backups. Besides the many practical problems you can have (laptop lost! fire!), SSDs have firmware bugs every so often.

SSD write fear is kinda nuts, IMO. It's a device subject to wear, tear, and catastrophic failure -- just like HDDs. Keep backups (three for important data; a cloud copy is only one copy), enjoy the performance, and don't sweat it. If you aren't keeping backups, then go ahead and cower under the sheets at night.

Since when do SSD's have a 100,000 write limit? My 256gb Samsung 840 has something like 2-3k, not 100k.