2

Ubuntu on a Crucial SSD, fantastic! But will never use 120 Gigs or even half that. After years of SSD writes (and subsequent SSD corruption), does an utility exist to copy the entire operating system and all files over to the unused portion of the SSD?

My understanding is the SSD degrades in activity areas, where it is being used and written to. But at least half of the SSD never used, and will never be used, and so remains in new pristine condition.

Can the lifetime of the SSD be doubled by copying the first 60 Gigs (where Ubuntu and everything resides) over to the last never-used 60 Gigs? Then those first 60 Gigs would be erased, and set aside as a new partition. And Ubuntu now runs on the equivalent of a new SSD, those last 60 Gigs. The utility would then need to prevent any future usage or writes to that first 60 Gig partition of the SSD where Ubuntu had formerly resided.

Improve this question

3 Answers 3

Reset to default
6

Solid-state drives are not like regular mechanical hard drives in the sense that you can partition the drive to keep data for certain partitions in a specific physical area of the drive platters. You cannot write to a specific memory address on the SSD, like you can with RAM.

The SSD controller will distribute data across all of the available memory on the drive. If it did not, the drive would quickly wear out specific blocks of memory, since each piece of memory can only be written to a few thousand times (depending on the type of memory). This feature is called wear leveling.

In your case, you are thinking about over-provisioning. Solid-state drives do this from the factory, you might have 256GB of memory chips on a drive, but only 240GB of it is available. The leftover memory is called spare area. The more spare area you have, the higher-performing and longer-lasting your drive will be. Of course, the trade off is that you end up with less storage capacity.

If you are never going to fill up your 120GB drive, you an do one of two things: First, just leave it alone and let the SSD controller take care of it. If you never fill the drive, it will use wear-leveling to distribute writes across all of the available memory. Second, you could partition the drive to say, 100GB, and then you would have 20GB additional spare area. In this case, you could fill the drive up and not worry about compromising it's performance because you have the extra 20GB of memory for the SSD controller to use.

Improve this answer
1

You can find a lot of information on the Easy Linux Tips Project. Basically what you call the "unused portion of ssd" may not really make sense as the system tries to use (or recycle) the whole disk in a periodic way so that each part gets a uniform number of writes in the long run. This is called wear leveling.

Improve this answer
0

Over the lifespan of your drive (not that you'll actually use it enough to achieve this), parts of it will become bad. Your operating system and SSD are capable of detecting these failures and preventing writes to bad blocks/sectors.

The best thing you can do is make sure your operating system and SSD are doing their jobs correctly.

Most people, when talking about SSD wear, over-think and create non-existing problems for themselves.

Improve this answer
1
  • 3
    I'm not sure, but as far as I know good SSD controllers do wear levelling in a way which is transparent to the OS. I.e. they move around the blocks as they appear to the OS on the disk, so when you use the first 60GB of the disk, internally those 60GB are sprawled all over the disk and change location regularly. Meaning you should just let the controller do it's job and stop worrying.
    – Nobody
    May 17, 2014 at 18:10

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .