On: StackOverflow Update: 560M Pageviews a Month, 25 S...

[ck2]

This is the fascinating part to me, their SSD have not failed:

Failures have not been a problem, even with hundreds of intel 2.5" SSDs in production, a single one hasn’t failed yet. One or more spare parts are kept for each model, but multiple drive failure hasn't been a concern.

[Nick-Craver → ck2]

Yep, still true. We lost one Intel 910 drive (PCIe SSD), and that was very abnormal - died so soon it was almost DOA. We hooked up directly with Intel for them to The replacement is still going strong as is another 910 we have.

All of those 2.5" Intels though, still trucking along! We're looking at some P3700s PCIe NVMe drives now, blog post coming about that.

[kmontrose → ck2]

There's a real difference between consumer and enterprise grade SSDs.

For comparison, for a while we nearly had a 1 to 1 dev to killed (consumer) SSD in work machines (see: http://blog.codinghorror.com/the-hot-crazy-solid-state-drive...). Granted, we did adopt them (see: http://blog.codinghorror.com/the-state-of-solid-state-hard-d...) for local dev use much earlier than we started putting them in servers.

I've killed 2 in my time here, though I do horrible big-ish data things that are either not SSD friendly or actively hostile.

[dekhn → ck2]

I would agree that hundreds of drives is a large enough sample that you would expect to see some failures.

That said, all the SSDs I've had have taken repeated pounding with no complaints. I accidentally swapped a few terabytes to swap on an SSD, and was simply surprised that the job I was doing finished faster.