If you’re in a panic because the Internet told you that your shiny new SSD may lose data in “just a few days” when stored in a hot room, take a chill pill—it’s apparently all a huge misunderstanding.
In a conversation with Kent Smith of Seagate and Alvin Cox, the Seagate engineer who wrote the presentation that set the Internet abuzz, PCWorld was told we’re all just reading it wrong.
“People have misunderstood the data,” Smith said.
Cox agreed, saying there’s no reason to fret.
“I wouldn’t worry about [losing data],” Cox told PCWorld. “This all pertains to end of life. As a consumer, an SSD product or even a flash product is never going to get to the point where it’s temperature-dependent on retaining the data.”
Why this matters: Users from New York to Rio De Janiero are freaking out over the risk of losing data when their SSDs are powered off. We decided to go to the source of it all for the truth.
The original presentation dates back to when Cox chaired a committee for JEDEC, the industry group that blesses memory specs. It was intended to help data center and enterprise customers understand what could happen to an SSD—but only after it had reached the end of its useful life span and was then stored at abnormal temperatures. It’s not intended to be applied to an SSD in the prime of its life in either an enterprise or a consumer setting.
But that’s not how the Internet viewed it. The presentation—almost five years old now—surfaced in a forensic computing blog as an explanation for why an SSD could start to lose data in a short amount of time at high temperatures. Once media outlets jumped on the story, it spread across the globe.
“The standards body for the microelectronics industry has found that Solid State Drives (SSD) can start to lose their data and become corrupted if they are left without power for as little as a week,” said the International Business Times, one of the first to run a story on the blog post. From there, the Internet seemed to amplify as fact that an SSD left unplugged would lose data—all citing Cox’s JEDEC presentation.
But Cox and Smith said that’s not correct. In fact, both said, an SSD that isn’t worn out rarely experiences data errors. Data center use also subjects SSDs to far more “program/erase” cycles than a typical consumer could under any normal circumstances.
Cox and Smith cited numerous tech websites that have torture-tested SSDs well beyond their rated lifespans using 24/7 work loads. The TechReport did manage to kill a number of SSDs, but only after writing hundreds of terabytes to them. Some of the drives still made it beyond the petabyte range.
Wear is one of the risk factors for SSD data loss at high temperatures, but because it’s nearly impossible for an average user to wear out an SSD, the danger is very small, Cox and Smith said. Even a worn-out SSD would still go a year without data loss, according to the original presentation, and that’s while being stored at 87 degrees Fahreneit the entire time.
For the same reasons, Smith said, enterprise customers are also unlikely to suffer from heat-related dead drive issues. Besides, they're more likely to use tape or other cheaper methods to back up data.
That’s not to say that SSDs aren’t immune from failures and data loss. Like all electronics, there’s always the risk of failure. Our own story helps put SSD failure rates in perspective.
Smith and Cox said the intent of the original presentation was to illustrate a worst-case scenario. What if the truck with the SSDs from the data center broke down, in the Arizona desert in July, on the way to the archiving center? How long could the truck be parked before data loss occurred from excessive heat? While that’s a scenario that could happen, it’s also highly unlikely—which is why the fear gripping SSD owners is unwarranted.