EnduranceAnalyzer won`t update

[KIlya]

Hi, we have a very high load server with ssd (intel ssd dc s3610 1.2 Tb). And we want to measure their life. We installed the latest Intel SSD DATA Center tool (3.0.6).

After command:

isdct.exe show -d enduranceanalyzer -intelssd X

We are getting for all our ssds:

EnduranceAnalyzer : Workload did not induce any wear on the drive. Expected life calculation is invalid. Ensure that workloads run for an hour or more for valid calculations.

Our ssds are in work for about 2 years. What kind of "workloads run for an hour"? How can we get enduranceanalyzer for our ssd?

We can't detach our ssds from our Production-server, as mentioned in Intel_SSD_Data_Center_Tool_3_0_x_User_Guide.pdf. And we already did the next command:

isdct.exe set –intelssd X enduranceanalyzer=reset

Still no result.

[idata]

Hello KI_medisacope,

We understand you would like to measure the life of your Intel® SSD DS S3610 Series.

Please keep in mind that the Endurance Analyzer is used to calculate the drive's life expectancy in years based on a certain workload (think "how long can I keep doing this before I get tired"). It does so by comparing certain SMART attributes before and after a test workload, then doing some maths.

For example (from the ISDCT User Guide 4.12, page 60):

1. Install the drive on a test system.

2. Reset the Endurance Analyzer.3. Apply the expected workload for at least one hour.4. Run the Endurance Analyzer command to display the drive's life expectancy in years based on previous workload.

If what you're looking for is the actual estimated life remaining of your drive, it's best to rely on the drive's SMART attributes directly.

To display the SMART attributes for one or more Intel® SSDs you'll need to run the show -smart command and look at the (E9) Media Wearout Indicator.

For SATA drives this attribute starts off with a normalized value of 100 which decreases down to zero as the memory blocks of your SSD reach their endurance limits and become worn out.

For more information on this command and usage examples please refer to the http://https//www.intel.com/content/dam/support/us/en/documents/solid-state-drives/ssd-software/Intel_SSD_DCT_3_0_x_User_Guide.pdf ISDCT User Guide, section 2.1.3, page 21 to 23.

For more information on this SSD's SMART attributes, please refer to it's http://https//www.intel.com/content/dam/www/public/us/en/documents/product-specifications/ssd-dc-s3610-spec.pdf Product Specification document, section 5.4.1, page 24.

We hope this information helps.

Best regards,

Carlos A.

[KIlya]

Yes. We exactly would like to know about the drive's life expectancy in years based on a our certain workload.

For example (from the ISDCT User Guide 4.12, page 60):

1. Install the drive on a test system.

2. Reset the Endurance Analyzer.3. Apply the expected workload for at least one hour.4. Run the Endurance Analyzer command to display the drive's life expectancy in years based on previous workload.

1. I already said that we can not do this, didn't i?
2. Didn't I say before that we already did this?
3. We apply workload on our prod servers several times. And more than at least an one hour.
4. Really? Did you even read what I said before? In my first message? We get every time the next message after reset and 1,2,12,24 hours

Workload did not induce any wear on the drive. Expected life calculation is invalid. Ensure that workloads run for an hour or more for valid calculations.

And we can't rely just on SMART E9. We need to know expected life.

[idata]

Hello KI_medisacope,

Could you please provided more information on how the drive is being used? Please keep in mind that reads don't add wear on your SSD. The Intel® SSD DC S3610 Series (1.2TB) is a high endurance data center model designed to endure up to 3 drive writes per day (1.2TB x3) 24/7 for 5 years (equaling 8.6 Petabytes).If used mostly for reads, with light write workloads, you may need to run the "enduranceanalyzer=reset" command, then apply your standard worload. If 24 hours doesn't do the trick, you may need to wait much longer before requesting the endurance analyzer results.Depending on your application, our Intel® SSD Endurance Calculator may also be helpful: http://estimator.intel.com/ssdendurance/Could you please provide us the following information for one of these drives:1. How long has this SSD been in use?2. Please share the SMART attributes for this SSD using: isdct.exe show –destination <</i>filenamepath> -intelssd <</i>device index> -smart  We look forward to hearing back from you.  Best regards, Carlos A.

[KIlya]

Good day Carlos A,

We made reset in friday and smart output now. Hope it will help.Our disks are in work for about 1,5 years.

Best regards,

Konstantin I.

[idata]

Hello KI_medisacope,

Checking the SMART attributes for your drive, we can tell you that the amount of host writes is not that high and for the purpose of the expected life of it, this one is fine.About the endurance analyzer, it may be required to let it run more writes before it can actually update the information; However, taking into consideration that you have already used the drive for 1.5 years, this SSD will definitely exceed the warranty period (5 years), as the SMART attributes look pretty decent.We would like to ask you: -What is the workload or how is this SSD being used?We'll be waiting for your response.Regards,Nestor C

[KIlya]

Good day Nestor C,

We are collecting data and analyzing them on production server with ssd.Can i ask then: what is the actual point of enduranceanalyzerm if it cant be updated everytime? When do it actually show the expected life?

Best regards,

Konstantin I.

[idata]

Hello KI_medisacope,

The Endurance Analyzer is intended to provide an estimated life expectancy for a drive under the actual workload that it will be subjected to. This is usually relevant for high write environments, where data is modified or added often. Since writing, modifying, or deleting data is the main source of wear, this allows system managers to be ready to replace the drive before the end of this period.Some SSDs on the other hand are used to store data that is accessed (but not necessarily modified) often. Since reading data places little to no wear on an SSD, this tool is not as helpful or necessary. In these cases we recommend monitoring the E9 Media Wear-out Indicator. This attribute starts off at 100, and slowly goes down to 0 before the drive can no longer be written to. Normally once this number reaches 20 or 10, you can start considering a replacement, however the SSD will still have plenty of life left over.Do keep in mind that a drive may still fail due to unexpected circumstances. Because of this, you may want to store your data in a redundancy RAID (such as RAID 5, 6, or 10), and always keep an updated backup.Best regards,Carlos A.

[idata]

Hello KI_medisacope,

We have not heard back from you in a while.If there is anything else we can do to help, please let us know.Best regards,Carlos A.