March 05, 2014 – St. Louis, MO – DataRecovery.com, a world leader in data recovery, warns computer users not to take Backblaze’s® hard drive failure statistics into consideration when making purchases.
Backblaze®, an online backup company, recently posted a comprehensive overview of various consumer hard drive models according to the company’s internal usage. The blog instantly made waves, as hard drive manufacturers typically post limited information about drive longevity.
“We don’t doubt the accuracy of Backblaze’s® information, but we don’t believe that their figures are representative of the industry as a whole,” said Ben Carmitchel, President of DataRecovery.com. “We still applaud Backblaze® for their invaluable service and for sharing that information to help their users, but their blog was never intended as a scientific analysis.”
According to DataRecovery.com’s media experts, Backblaze’s® small sample size could affect their results, and is contradictory to Datarecovery.com’s internal analysis which samples a far greater range of hard drives –all of which failed due to many different circumstances. Backblaze® reported information for more than 25,000 individual hard drives, and the most popular model in their pods was Seagate’s® ST4000DM000 at 5,199 drives.
However, some of the reporting used significantly smaller sample sizes. Backblaze® displayed failure rates for less than 1,000 of another manufacturer’s hard drives, and for two hard drive models, Backblaze® used less than 100 drives. This contributed to a notable failure rating of 120 percent for another Seagate® model (Backblaze® included warranty replacement drives in their numbers, which allowed failure rates to exceed 100 percent).
“Hard drive manufacturers ship millions of drives every day from many different factories, and a sample size of less than 2,000 or so will result in extremely inaccurate averages,” said Carmitchel.
“Even for those larger samples, Backblaze® would need to show that the drives’ failure rates were representative, and that’s a big issue. For instance, if a large number of the drives came from the same factory or if they were made within a short span of time, that would skew the statistics. That’s not even touching on drives’ working environment.”
More importantly, DataRecovery.com’s experts believe that hard drive buyers might be asking the wrong question. All major hard drive manufacturers have strict standards for acceptable failure rates, and according to DataRecovery.com’s internal analyses, prominent brands are very competitive in terms of longevity.
There are far too many variables involved to predict when a particular drive will fail, and Carmitchel says that buyers should look at other features such as seek times, noise levels, and power usage.
“It’s important to remember that every hard drive will fail at some point, and rather than trying to predict the failure, computer users should back up their data regularly and take appropriate actions to minimize stress on their systems,” Carmitchel said. “In the long run, it’s a much better approach.”
Datarecovery.com is a world leader in data recovery and computer forensics. With four locations in California, Illinois, Arizona, and Toronto, the company provides a variety of services to thousands of clients each year. Visit www.datarecovery.com for more information.
Ph. 800.237.4200 ext. 269