Backblaze, one of the largest cloud storage companies, has released its quarterly HDD stats, which cover crucial aspects such as failure rate. The results obtained are through detailed testing with a huge sampling size.
Backblaze's New Report Reveals AFR Seeing An Increment, 8TB & 10TB HDD Models Being the Primary Suspect
The total HDDs (Hard Disk Drives) tested by the company reached an astonishing 241,297 hard drives. The actual number was a bit higher than this since the company removed 357 drives as they were utilized for testing. The data compiled accounts for a total of 240,940 hard drives arranged in 31 different groups depending upon manufacturer and results obtained.
Analyzing the results, it can be seen that the AFR (annualized failure rate) for this quarter was 2.28%, which is a noticeable jump from the previous quarter's AFR of 1.54%. While the author expresses AFR results as "volatile", the increase has also put the company in doubt. The upwards trend of AFR is mainly due to the aging drives since the higher storage ones such as the 8TB and 10TB HDDs aren't holding up to expectations.
Diving into the interesting bits, Backblaze mentions that the oldest average drive they have up and running is a 6TB Seagate (model: ST6000DX000) which has an uptime of 8.2 years. The oldest operational drive is the 4TB Seagate (model: ST4000DM000), reaching 8.8 years. Moreover, when it comes to zero failures, Backblaze identified several models from companies like Seagate and Toshiba, running at least 50,000 drive dates, before reaching the "zero failure" milestone.
Lifetime HDD Failure Rate Also Witnesses Increase, Here is Why
We mentioned earlier that the 8TB and 10TB drives were the main culprits behind the massive quarterly AFR hike. Backblaze analyzed the Lifetime AFR to support the claim, seeing that it has also experienced an increase of 0.05%. Through its testing, the company found that the 8TB and 10TB HDDs are the leading cause for the higher AFR.
You can see a chart below which lists the tested storages, the "red' baseline indicates the AFR, all values above contribute to the increase, and all values below are subtracted from the values of AFR.
The company didn't specify the reason behind this pattern, but several probable causes could be considered, such as temperatures, drive quality, etc. Dropping a note for cloud companies, the author Andy Klein assures the readers, claiming that "drive failures" have become a norm in the industry. Here is what he has to say:
Are we worried about the increase in drive failure rates? Of course we’d like to see them lower, but the inescapable reality of the cloud storage business is that drives fail. Over the years, we have seen a wide range of failure rates across different manufacturers, drive models, and drive sizes. If you are not prepared for that, you will fail.
Andy Klein via Backblaze
Many of us surely think about the motive behind this extensive research, especially on "HDD failures." Backblaze, apart from being a leading cloud computing firm, puts in the effort to provide such data, which not only helps manufacturing companies but also aids an average consumer and even a commercial buyer. It involves a lot of work, and we won't forget to appreciate the company publishing such detailed data.
News Source: Backblaze
