Welcome to the MacNN Forums.

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

You are here: MacNN Forums > News > Mac News > MacNN launches testing of heat on unpowered SSD data retention

MacNN launches testing of heat on unpowered SSD data retention
Thread Tools
NewsPoster
MacNN Staff
Join Date: Jul 2012
Status: Offline
Reply With Quote
Aug 31, 2015, 01:01 PM
 
Earlier this year, there was some controversy about heat and SSD data life when left unpowered. We don't think this is a major problem for most users, as a SSD will be powered for essentially its entire operational life -- but it may be an issue going forward, as users collect older technologies for re-use, given the generally long life of Apple hardware. So, in conjunction with our SSD Trim testing, we've started real-life testing of unpowered, depleted SSDs, exposed to high temperatures for extended periods of time to check on both data retention, as well as drive survivability.

Materials

We've been given access to a temperature- and humidity-controlled chamber. The temperature has been set to 160F, and 50 percent humidity for the duration of the test. We've been given seven 250GB Samsung 840 EVO SSDs that have all seen 600TB or more of data copies, and been taken off service. Eight others of the same make and model are factory fresh.

We've formatted all the drives, and are using a 200GB subset of the data we've been using for the Trim testing. Six of the depleted drives, and six of the factory-fresh drives, are already headed to the test chamber later today. One pair of depleted and fresh drives are sitting in a cabinet near our home server, and will remain untouched for the full year as a control.

As an additional control, we've taken one of the new drives with the same data, and installed it into a USB 3.0 external forensics case with ventilation, preventing writes by the OS or accidental erasure, and connected it to our home server where it will remain powered but unaltered for the duration of the test. We also have two other copies of the data on archival media in the case of an accident or failure of the SSD.

Testing methodology

This is a long-term test. Every two months, we'll pull out one factory-fresh drive, and one older drive. We'll then install them both in forensics drive cases, preventing writing to the drive by the OS, and compare checksums to our control data on archival media as well as the powered SSD.

We'll repeat this process every two months for a year. Additionally, we'll put the drives we pulled out of climate control for testing and comparison back in and pull them all out in two years, for a second year of data -- the first drive we test will have been in the 160F environment for 22 months at that point after reinsertion.

Potential Pitfalls

The test won't be completely conclusive. The effect of cosmic rays is somewhat lessened in the climate chamber. Radiation levels inside the test chamber is 7.1 µrem per hour (0.071 µSv), and inside the case connected to the home server is 8.2 µrem per hour. Both of these numbers are slightly less than the background radiation of 10.4 µrem per hour in the location of the test chamber, and 9.8 µrem per hour at the home server's location (variances due to building materials, in all likelihood). We're working with our test partners on drive radiation exposure testing, but we're not ready to start that yet.

Furthermore, the testing is being performed on Samsung's TLC SSD media -- not universal to all manufacturers. However, other manufacturers now have similar processes, and we expect that it will be more widely adopted going forward.

So, come back every few months. We'll see where this takes us.

-Mike Wuerthele
Managing Editor
( Last edited by NewsPoster; Aug 31, 2015 at 01:14 PM. )
     
chimaera
Dedicated MacNNer
Join Date: Apr 2007
Status: Offline
Reply With Quote
Aug 31, 2015, 01:59 PM
 
In the event of checksum failures, I assume you'll do a full bit-compare, to determine the exact failure rate?

Suggestion: throw a few mid-life SSDs into the mix. If the old ones fail heavily while the new ones hold up, there will be little useful data without midpoint plots on the graph.
     
Mike Wuerthele
Managing Editor
Join Date: Jul 2012
Status: Offline
Reply With Quote
Aug 31, 2015, 02:20 PM
 
Originally Posted by chimaera View Post
In the event of checksum failures, I assume you'll do a full bit-compare, to determine the exact failure rate?
You betcha.

We thought about mid-life, and it can be argued that we don't even know what "life" is. While we're not trustful of the drives we've used for the Trim testing, other research suggests that drives can get out to petabytes of transfers before any real serious issues arise. We'll see how it goes. We've got access to the chamber for a few years, and we're about at the limit of reseller and vendor support for the drives as it is.

And before the next question about radiation measurement crops up, eight years as a chemist and radiological technician in the USN Submarine fleet qualifies me
     
chimaera
Dedicated MacNNer
Join Date: Apr 2007
Status: Offline
Reply With Quote
Aug 31, 2015, 02:49 PM
 
Heh, you're safe. I wasn't going to ask about radiation. Natural background varies by location, and it's not subject to user control anyway. Not many people buy that book on how to radiation-proof your home with concrete.
     
Feathers
Grizzled Veteran
Join Date: Oct 1999
Location: South Pole
Status: Offline
Reply With Quote
Sep 1, 2015, 11:46 AM
 
As far as radiation is concerned, the only issue is exposure to x-rays in airports etc. It has been confirmed that x-rays can flip a zero to a one but not vice-versa because of the way SSD memory works. The fact that data can be corrupted in this way is of concern but obviously not the subject of this test. Great to see it getting done and will be interested in the results.
     
   
Thread Tools
 
Forum Links
Forum Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Top
Privacy Policy
All times are GMT -4. The time now is 07:40 AM.
All contents of these forums © 1995-2017 MacNN. All rights reserved.
Branding + Design: www.gesamtbild.com
vBulletin v.3.8.8 © 2000-2017, Jelsoft Enterprises Ltd.,