SSD? To optimize OR not to optimize that is the question
 

Hi

If you have a SSD drive many distributions normally treat it like a normal hard drive during installation -- thus giving no SSD optimization for extended life span. I was looking at at these two articles on SSD preparation before and after install:

Ubuntu, Mint, Debian on SSD
https://sites.google.com/site/easylinuxtipsproject/ssd

Freebsd on SSD
http://www.wonkity.com/~wblock/docs/html/ssd.html

Wouldn't it be great if distro developers had an option built into the installation wizard to choice if you will install linux to a SSD or Hard drive? And if you choose the SSD drive option, the installation wizard would do all the SSD optimization for you and your applications.

When I install linux on my SSD drive I didn't know it needed to be optimizes to prolong the lifespan of my SSD. To be honest, I treated my SSD as a magnetic hard drive.

So, do you optimize your SSD drive or not?

Also, do you feel the distro developer should integrate an option to choose a SSD drive or hard drive during installation that way it can optimize the SSD for you? Or do you prefer the DIY route ( Do it yourself )?

Thanks, looking forward to your opinions :-)

it was discussed here earlier (actually I can't find the thread), the result was: it is not really important, but if you have 5-10 minutes to improve why not?

For me "optimizing" my SSDs comes down to using a filesystem that supports TRIM and running the fstrim command from time to time.
Wear-out due to writing to the disk is something that can be neglected on modern SSDs, so I am not concerned at all about that.

Quite honestly, I'm gonna wait for SSD technology to really mature. Today's stuff smacks too much of "bubble memory" for my personal taste. :)

(Bah. As a matter of fact, I do remember bubble-memory! What's it to ya, kid?) ;-)

I know that, in a few years' time, these devices won't "wear out when you write to them too much," and so on. They're being marketed, of course, and installed in products that are being profitably sold to consumers, but instinct tells me that if you stay away from bleeding-edges, you don't get cut nearly as often. My next laptop will have an HDD in it. Three years from now, when I buy my next one (I buy extended warranties and swap machines when the warranty runs out ...), it will probably have a SSD, and that device will by then be vastly-improved. I can wait.

SSDs are pretty good already and wear-out has already become a npn-issue. If you want to have a look at an endurance test you may want to have a look here: http://techreport.com/review/26523/t...-to-a-petabyte
While all tested SSDs worked fine after 600TB written data on their way to the PetaByte mark the first disks stopped working.
The SSD in main machine is now at about 19000 hours power-on time (used for gaming with Windows and Linux, both OSes on the SSD, also for software compiling), but "only" at about 3.5TB written data. At least for me it seems that the SSD will have a long lifetime.

I said the same thing, however I do realize the need to keep up with the industry trends. So I bought my first SSD a few months ago and a "new" laptop to stick it in. What worked out for me was a sync system where as soon as my data is written its synced up to my server. So pretty much anything useful in my /home is always copied elsewhere should the SSD die. So now I can run the SSD and test it for a while before committing to it.

According to smartctl, I have nearly 24,000 hours on my SSD, and have only used 4% of it's life. This is what I have in my /etc/fstab:

Can't remember if I did this, or if this was default (Firefox - about:config):
I'm sure there's more I could do. If I've only used 4% of it's life in the last 3 years, then chances are I will probably have upgraded before the SSD reached EOL.