[Discuss] SSD drives vs. Mechanical drives

Edward Ned Harvey (blu) blu at nedharvey.com
Tue May 6 07:22:25 EDT 2014


> From: discuss-bounces+blu=nedharvey.com at blu.org [mailto:discuss-
> bounces+blu=nedharvey.com at blu.org] On Behalf Of Kent Borg
> 
> I don't trust flash, so I am being cautious, but...

Everything you said is equally true of rust.  

HDD's can also die with no warning and no recourse.  The question would be:  What's the probability, and what's the value?  If value is sufficient, then you add redundancy in the form of mirroring, etc.  The probability of a USB stick failing is very high.  But the probability of SATA SSD failing is on-par with HDD.

Don't know where you get the idea flash hates writes.  The honest truth is each individual cell will degrade slightly each time it's written, and on USB devices, they will fail quickly due to lack of wear leveling.  But any SATA SSD should include wear leveling, which means, only after you have completely overwritten the entire contents of the drive somewhere between 10,000 and 100m times or so, then it becomes relevant.  (250G drive, after you write a few petabytes or so.  How many years does this take?)  It's simply not a real-world issue.

SSD's and HDD's both hate getting full.  The grain of truth that you're alluding to is that the flash drive requires some empty space to work with, for memory management, block recycling and consolidation, etc.  But the drive reserves a percentage of its capacity, not presented to the OS, so it's always available.  Incidentally, the same is true for HDD's, just on a smaller scale.  When the HDD finds some checksum failures, if it's able to recover and mark it as a soft fail, then the HDD will remap bad sectors, and it reserves a percentage of the drive, not presented to the OS, to satisfy those situations.



More information about the Discuss mailing list