• Welcome to iHelpForum - the place to get help from knowledgeable techs in all areas of Tech, Home and Auto help. Consider checking out our Guides or Registering an account to post on our forums today.

Conversation Not so solid state of affairs


Because, you know, Obama.
I just bought a 250GB EVO SSD. I do not need a 250GB SSD. It was a bundle deal.

If I partition, is wear leveling performed across the partition or the entire disk?

If partition, does that mean I could double the life of the device by part'g the device in two equal halves (and leave the second half raw)?


Beware of the Bear...
iHF Veteran
WCG Team Member
I do not know either.

I only have had a 60 gig SSD that run out of room quick smart, by the time Winders grabbed two chunks of space for the swap paging file (I have 16gigs of ram, and let winders manage it for the paging file), and hiberfil.sys.

I put a 500 gigs Samsung harddrive with two partitions of 250 gigs in and got rid of the ssd completely

Leave the SSD as it is I reckon...
Last edited:


In at the Deep End...
Security Advisor
iHF Legend
The read/write count is constant across the whole drive as with any flash type memory. Now I'm sure there will be some troll out there that will call me an old fool and they would be right but not in this case I think. ;)
I guess that answers that then...


Because, you know, Obama.
The read/write count is constant across the whole drive as with any flash type memory...
What counter? That lead me to some MS sources and then to Micron (pdf link).

Difficult to find the right words, especially since I am still wrapping my head around the subject.
*Cumulative volume of writes, not cumulative count of writes. Total Bytes Written

All drives are virtual. In terms of addressing, Windows does not distinguish SSD vs HDD.
The drive's firmware does not see partitions, therefore, wear leveling occurs across the entire cell set.
This is close enough to correct to justify the suppositions in Cameldung's link.

As it turns out, the answer is device specific. At least controller specific. I am looking for whitepapers on the MGX controller.

My interest in this has nothing to with a system drive. I am considering moving a 6TB archive to SSD. I mirror my archive, so I need 12TB of SSD. The mirror is rarely powered on - only to re-mirror.

The choice I am wrestling with: should I RAID0 six 1TB SSDs? If yes, how often should I replace disks?

I got 7 years out of some WD 500GB disks. Sadly, I got less than 3 years out of the variety of 1.5TB that replaced them.

Complicating matters - NLT than Christmas '15, 6TB disks will retail at $150.


Because, you know, Obama.
Arctos grossly misconfigured his 60GB SSD, but I ain't gonna tell him.


Because, you know, Obama.
I take a lot of home video. A lot. Much. Tons. I am too lazy to edit it. Way too lazy.

I want this content to be accessible in 2050 AD. Long after I am dead.


Because, you know, Obama.
From the pdf linked above:
"These prematurely worn blocks (or entire devices) can cause the SSD to operate as read-only before all blocks and/or devices are fully used."

Does that mean (it is possible) that existing data has a stable shelf life as long as the SSD is electrically viable? What happens when an SSD has exceeded its wear leveling lifetime?