Armand

Things that will damage SSD

Recommended Posts

What are the things done on the SSD that will eventually damage the SSD? Let us consider that everything is used in a normal method.

To ask this in another perspective, what are the common problems that SSD have?

Share this post


Link to post
Share on other sites

Really, writing abusive amounts of data to the drive is about all that "hurts" it. Running defrag and other things can wear it out too I suppose. When we test drives we generally write a lot of data, but even that only eats up a small percentage of its useful like.

Share this post


Link to post
Share on other sites

I agree with Brian. Also, I think that writing, erasing, and rewriting data cycles would definitely wear out an SSD. I'm not sure what else would, though.

* any unnecessary writes reduce the lifespan.

* Formatting it to the full advertised capacity will reduce the lifespan of the SSD. See http://intelstudios.edgesuite.net/idf/2009/sf/aep/IDF_2009_MEMS003/f.htm

The video has a 30+ minute presentation. Only the first 15 minutes is worth listening to but it has solid data and concepts pertinent to reliability of consumer grade SSDs.

Reducing the partition size can increase the life of the SSD by over 3x as in over 300% or over 200% increase depending on how you like to think about the end result or the increase.

Unfortunately this is only true if you have never written to the drive's full capacity. A secure erase is the only way to return the drive to factory fresh and allow a smaller partition to increase the drives lifespan. Different drive controllers will deal with this better or worse but generally unpartitioned space will improve the longevity of a SSD.

After that you get into

* extreme heat such as the interior of a car in Arizona or a server in a room with tons of servers and no air conditioning (>70c/158f)

* extreme cold such as used in some overclocking experiments (don't spill your liquid nitrogen on my SSD)

* large electrical fields that could cause shorts or burn out a transistor with transient current (don't place it inside medical imaging equipment).

* use of vice, saw, explosives, other blunt forces (oh, that video with the 3wood golf club busting up a SSD is a fine example)

Share this post


Link to post
Share on other sites

That IDF session is a good watch for any first time SSD owner. It's pretty amazing how much endurance is gained by giving up a little capacity, seems a fair trade.

Share this post


Link to post
Share on other sites

Dhanson865, I really appreciate the tutorial or precaution that SSD owners should do to lengthen the lifespan of the SSD. I've never thought that formatting will lessen the life of an SSD.

Share this post


Link to post
Share on other sites

Will it have the same effect if we format the HDD often. I know this should be discussed in the HDD forum but the frequent formatting was already mentioned here.

How can we maintain the optimum performance of the SSD then?

Share this post


Link to post
Share on other sites

Will it have the same effect if we format the HDD often. I know this should be discussed in the HDD forum but the frequent formatting was already mentioned here.

How can we maintain the optimum performance of the SSD then?

Hard drives in most cases won't care if you do or don't format them often, once, full, partial, doesn't matter. There are extreme cases if you need to store a hard drive for legal purposes or historical purposes but end users shouldn't care.

To maintain the proper performance of an SSD you keep the firmware updated and follow the manufacturers suggestions. If you have an Intel G2 drive they have a toolkit that can be run to make sure all is good. If you don't have an Intel SSD you have to learn the specifics of your SSD and do what is needed in other ways.

Share this post


Link to post
Share on other sites

Will it have the same effect if we format the HDD often. I know this should be discussed in the HDD forum but the frequent formatting was already mentioned here.

How can we maintain the optimum performance of the SSD then?

SSDs will also include SMART data just like a HDD does so you can gain the data from any program as to how your hard disk is doing at that point. If it is noticing any slides in performance it should let you know and tell you how to deal with it.

Share this post


Link to post
Share on other sites

I'm sure an electromagnetic pulse can kill it... but unless you plan to use your SSD in the upper atmosphere near nuclear explosions I think you are quite safe from those :D

In normal life unless you are setting out to kill one pretty much nothing should - although you possibly should be careful around induced currents of any form.

Share this post


Link to post
Share on other sites

I'm sure an electromagnetic pulse can kill it...

Yes, especially those coming out of high powered lasers, preferably pulsed ones. Rule of thumb: anything that causes ablation on stuff of similar color (piece of plastic or paper) should be avoided for prolonged amounts of time. You should definitely stop once you fired through the SSD casing and are starting to remove the chips plastic casing.

MrS

Share this post


Link to post
Share on other sites

RANDOM WRITES kill ssd exponentially faster than sequential writes.

note the differnce in write capacity...random V sequential.

idf.png

intel.png

Share this post


Link to post
Share on other sites

That is from Intel. one is from an IDF presentation and the other is from the user manual appendum for X-25M (they didnt release these specs until the drives were already out).

The reason why is complex, but basically during a small file random write one 4k file can occupy a entire block on the ssd. write combining does not work as effectively during random writes, magnifying write amplification.

You can kill a ssd in days/weeks if you just ran Iometer with a high Queue Depth random write test.

Edited by Computurd

Share this post


Link to post
Share on other sites

That is from Intel. one is from an IDF presentation and the other is from the user manual appendum for X-25M (they didnt release these specs until the drives were already out).

The reason why is complex, but basically during a small file random write one 4k file can occupy a entire block on the ssd. write combining does not work as effectively during random writes, magnifying write amplification.

You can kill a ssd in days/weeks if you just ran Iometer with a high Queue Depth random write test.

Well you did a good start :D

That a small file can occupy one full 4K block makes sense :)

Edited by DetlevCM

Share this post


Link to post
Share on other sites

Intel kept touting as a unique feature of Intel SSDs that they could use unpartitioned, never written (or secure erased) space as extra "spare" space. So would this not help on a SandForce, Indilinx or Toshiba based drive? Can anyone confirm either way?

Share this post


Link to post
Share on other sites

Intel kept touting as a unique feature of Intel SSDs that they could use unpartitioned, never written (or secure erased) space as extra "spare" space. So would this not help on a SandForce, Indilinx or Toshiba based drive? Can anyone confirm either way?

Intel might do a better job of it but any drive with wear leveling will take advantage of such space. Also sandforce beats that issue from another direction by reducing writes before they happen instead of managing the free space so well.

Share this post


Link to post
Share on other sites

Considering this thread has been revived, some data from my 160GB G2 Intel SSD:

Writes 9,22TB

Reallocated Sector Count 5

-> Otherwise still working well :) (Just like new)

Share this post


Link to post
Share on other sites

The link from intelstudios doesn't work.

* any unnecessary writes reduce the lifespan.

* Formatting it to the full advertised capacity will reduce the lifespan of the SSD. See http://intelstudios.edgesuite.net/idf/2009/sf/aep/IDF_2009_MEMS003/f.htm

The video has a 30+ minute presentation. Only the first 15 minutes is worth listening to but it has solid data and concepts pertinent to reliability of consumer grade SSDs.

Reducing the partition size can increase the life of the SSD by over 3x as in over 300% or over 200% increase depending on how you like to think about the end result or the increase.

Unfortunately this is only true if you have never written to the drive's full capacity. A secure erase is the only way to return the drive to factory fresh and allow a smaller partition to increase the drives lifespan. Different drive controllers will deal with this better or worse but generally unpartitioned space will improve the longevity of a SSD.

After that you get into

* extreme heat such as the interior of a car in Arizona or a server in a room with tons of servers and no air conditioning (>70c/158f)

* extreme cold such as used in some overclocking experiments (don't spill your liquid nitrogen on my SSD)

* large electrical fields that could cause shorts or burn out a transistor with transient current (don't place it inside medical imaging equipment).

* use of vice, saw, explosives, other blunt forces (oh, that video with the 3wood golf club busting up a SSD is a fine example)

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now