It is worth noting that Apple, Micromat (TechTool Pro), and Prosoftengineering (Drive Genius) have all removed surface scans from their applications (granted Micromat and Prosoftengineering followed Apple's not too subtle lead) because in the opinion of their engineering staffs surface scans of SSDs could result in reducing the usable lifespan of those drives.
In the case of a Fusion drive the idea of scanning only the rotating rust segment is intuitively appealing, but the fusion takes place at a very low level within the logical structure of the volume and might require breaking that fusion to access only the rotating rust portion of the drive. In my experience it is possible to separate the two drive segments, but the process is destructive so all the data on the fused drive is lost. The process would look something like… (Omitted for brevity)
That's educational; thanks.
Considering the vulnerability of HDDs and the major role they play in fusion drives, you'd think that Apple would provide a means to do a surface scan.
Which is why I'd love for V1 to get involved in this conversation; he's apparently quite adept at thinking his way through problems that leave others tearing their hair out. (He hasn't posted in more than 3 months...very unusual, and as I said, I hope he's OK.)
Food for thought, though... This post by MMT3 (When's the last time I typed that?), albeit 6 years old makes what I think is a valid point.
The name of the Surface Scan test is not particularly apt for SSD drives, but the test is still valid. It is simply reading blocks, one after another, to make sure that they can be read. We have had customers run the Surface Scan to check SD cards used in cameras, and a few have been found to be defective.
Save the results of the Surface Scan, so it a later lest finds errors (unremapped bad blocks), you will have a good basis for a warranty claim.
It seems as if running a surface scan on an SSD when it's brand new and then again periodically but infrequently is a good idea.
And as for its shortening the drive's life, I think it was tacit who did some math a few years back and demonstrated that unless you're doing an
immense amount of graphics editing, your computer will die long before your SSD does...a few surface scans notwithstanding, I assume.
