Different threshholds for different drives

[soyo]

Hi hdsentinel

I installed today the newest version, and I got a "suprise". My SSD that I bought a year ago, suddenly showed 88% health due to "Wear Leveling Count", which I guess is something you changed in determining over the past couple versions.

Anyways my question regards the threshholds. Is it possible to set different threshholds for different drives? Ideally, the values that are now in the general threshholds tab I would like to be assigned to all the drives UNLESS there are some individual values set for any drive. For example maybe I would like to push higher a Bad region for my third drive to say 75% (just an example)

If there is no such feature, would you consider implementing it in the next version?

cheers
Lucas

[hdsentinel]

Hi Lucas,

SSD models use special, vendor specific attributes to determine the overall health of the memory cells of the SSD.
Even if no problems, errors reported by the SSD, the writes (because the memory cells tolerate only limited number of overwrite cycles) slowly but constantly reduce this value - and it is used to determine the overall health of the SSD.
Older Hard Disk Sentinel versions may not be able to determine newer SSD models and their such special values - this is why the newest version (for newer SSDs) may provide this value more accurately.

Thanks for the suggestion!
Currently only the temperature thresholds can be set individually for the drives, but I agree that it may be good idea to specify different health thresholds for the yellow (warning) and red (alert) levels.
Currently it is possible in Hard Disk Sentinel Enterprise version only, but it is possible that it will be available in the Professional version as well.

[soyo]

Hi Lucas,

SSD models use special, vendor specific attributes to determine the overall health of the memory cells of the SSD.
Even if no problems, errors reported by the SSD, the writes (because the memory cells tolerate only limited number of overwrite cycles) slowly but constantly reduce this value - and it is used to determine the overall health of the SSD.
Older Hard Disk Sentinel versions may not be able to determine newer SSD models and their such special values - this is why the newest version (for newer SSDs) may provide this value more accurately.

Thanks for a detailed answer. That was probibly it.

Can I ask, what is the border value of this SMART reading? I have Samsung 830 SSD btw.

Thanks for the suggestion!
Currently only the temperature thresholds can be set individually for the drives, but I agree that it may be good idea to specify different health thresholds for the yellow (warning) and red (alert) levels.
Currently it is possible in Hard Disk Sentinel Enterprise version only, but it is possible that it will be available in the Professional version as well.

Ofcourse I can only ask you for it, as it depends on your business plan. But it really is the only thing missing for me. So again, please consider it in the near future.

cheers
Lucas

[hdsentinel]

> Can I ask, what is the border value of this SMART reading? I have Samsung 830 SSD btw.

It will decrease during the lifetime of the SSD - even low to 1%.
Lower health value may be an indicator that the memory cells can't properly hold the stored information.
It may be still used - but the risk of data loss / corruption may increase as the health decrease.
Personally I'd recommend to consider replacement under 25% health, when the health goes below the red level.

> Ofcourse I can only ask you for it, as it depends on your business plan. But it really is the only thing missing for
> me. So again, please consider it in the near future.

Thanks for the tip, yes, I can make sure it will be available.

If you can use Report -> Send test report to developer option, that may help (and if you prefer, it is possible to provide test-version as soon as possible).

[soyo]

It will decrease during the lifetime of the SSD - even low to 1%.
Lower health value may be an indicator that the memory cells can't properly hold the stored information.
It may be still used - but the risk of data loss / corruption may increase as the health decrease.
Personally I'd recommend to consider replacement under 25% health, when the health goes below the red level.

Ok, so to translate it into values: The value this attribute shows for my drive is 435, and it also gives % value, which is 88. Lets approximate it just for the sake of simplicity, and say 400 would relate to 90%. That means, by the time it will reach 0%, the value will be around 4000. I'm asking about it, because I saw smart values from ssd owners, that had this value in the range of 20000+.

Could you also elaborate a bit on what happens in effect, when "wear leveling count" reaches 0%? I mean what does it actually mean for the SSD in it's operation, leaving aside the technical details?

Also, how does this attribute for the SSD relate to the Reallocated Sector Count?

If you can use Report -> Send test report to developer option, that may help (and if you prefer, it is possible to provide test-version as soon as possible).

Done

cheers
Lucas

[hdsentinel]

> Ok, so to translate it into values: The value this attribute shows for my drive is 435, and it also gives % value, which is 88.
> Lets approximate it just for the sake of simplicity, and say 400 would relate to 90%. That means, by the time it will reach 0%,
> the value will be around 4000. I'm asking about it, because I saw smart values from ssd owners, that had this value in the range of 20000+.

Yes, exactly as you wrote.
Also it changes (approx. linear way) as the amount of data written to the SSD.

> Could you also elaborate a bit on what happens in effect, when "wear leveling count" reaches 0%?

It stops at 1%, will not decrease below that.

> I mean what does it actually mean for the SSD in it's operation, leaving aside the technical details?

The SSD is still usable, can be used. However, more and more times it stops for longer time during access, the overall performance of the SSD (both reading and writing) drops drastically.


> Also, how does this attribute for the SSD relate to the Reallocated Sector Count?

It does not really affect Reallocated Sector Count directly. Yes, we may feel that (due to the wearout of the memory cells) more and more sectors will be reallocated, but this does not happen all the time.
But data corruption may occur if the SSD is not used all the time (then the memory cells may not be able to hold the stored information as the data can't be refreshed periodically be the dynamic wear leveling).

[soyo]

The SSD is still usable, can be used. However, more and more times it stops for longer time during access, the overall performance of the SSD (both reading and writing) drops drastically.

So this factor doesn't corrupt data as such, but just "slows down" the SSD? Is that a fair statement?

cheers
Lucas

[hdsentinel]

Please read the end of the answer also There I describe that data loss may occur as well if the memory cells can't retain the stored information.

[soyo]

Alright. Thanks for all the answers hdsentinel.

I will keep this thread as marked, so if there will be any news regarding the threshold functionality, please let me know.

cheers
Lucas

[hdsentinel]

Of course, I'll let you know any progress in this field.

[soyo]

Hey hdsentinel

Any news regarding implementing this feature in the PRO version?

cheers
Lucas

[hdsentinel]

Hi Lucas,

This feature is scheduled to be developed - but it is not yet available. Hopefully in the near future it can be added.

[soyo]

Thanks for instant reply hdsentinel.

Any vague plans as to when you plan next update?

cheers
Lucas

[hdsentinel]

I can't really promise an exact date - it is scheduled and will be available in a later version.

[soyo]

I can't really promise an exact date - it is scheduled and will be available in a later version.

Awsome. Well, patience is gold, and I am a patient man

Looking forward to this option in the next update and thank you for prompt response.

cheers
Lucas

[soyo]

Hi hdsentinel

Any update after a year? As you see I'm indeed a patient man

cheers
Lucas

[hdsentinel]

Hi Lucas,

Thanks for your patience

Sorry for the delay, hardware compatibility and constantly adding support for new devices took excessive amount of time and resources, but this is the most important. This may delay some of these new features.

Sorry, at the moment, it is not yet available.
Maybe later, in a future version (originally it was planned for 5.00, so it is possible that it will be available then).

[soyo]

Thanks for the update, I appreciate it.