|
|
Post by Boomzilla on May 28, 2020 14:04:37 GMT -5
I'm currently running a 2013 MacBook Pro with a factory SSD drive. One of the aftermarket programs I installed is called "Disk Drill." It popped up an alarm on my bar this morning saying:
APPLE SSD SM0512F
SSD Wear Level = 88%
Uptime = 16,060 hours
Power Cycle Count = 128,380
The disc is still working reliably. Should I swap it out now or run it to destruction?
Thanks - Boom
|
|
|
|
Post by vcautokid on May 28, 2020 14:24:26 GMT -5
|
|
|
|
Post by monkumonku on May 28, 2020 14:40:03 GMT -5
I'm currently running a 2013 MacBook Pro with a factory SSD drive. One of the aftermarket programs I installed is called "Disk Drill." It popped up an alarm on my bar this morning saying: APPLE SSD SM0512F SSD Wear Level = 88% Uptime = 16,060 hours Power Cycle Count = 128,380 The disc is still working reliably. Should I swap it out now or run it to destruction? Thanks - Boom You can afford it. Why take a chance? That's what that warning is for! |
|
|
|
Post by AudioHTIT on May 28, 2020 14:44:15 GMT -5
I have the same age MacBook Pro with that SSD, I've also not had any trouble and hadn't thought about replacing the hard drive. However I keep most of my data on an external SSD that I can move between my Mac mini(s) and the MacBook, so the internal drive is mostly OS and workspace. I keep them all backed up.
I've never heard of Disk Drill, but my plan is to use the drives until they fail, or for some other reason I want something different.
|
|
|
|
Post by Gary Cook on May 28, 2020 14:54:31 GMT -5
I need some help here, how does an SSD wear and how is any wear measured? In a spinner the rpm is measurable, as are the read and write speeds and the dead sectors can be counted. But an SSD, what is measured to determine wear?
Cheers
Gary
|
|
|
|
Post by adaboy on May 28, 2020 14:58:02 GMT -5
Agree totally with keeping it backed up and running until it fails, if it fails. "While you still own it" |
|
|
|
Post by garbulky on May 28, 2020 15:14:40 GMT -5
I would.
|
|
|
|
Post by Boomzilla on May 28, 2020 15:25:43 GMT -5
What wears out on a solid state drive is the magnetic flux of the memory array. Like a USB stick, the physical memory locations are designed for a fixed number of read/write cycles before the chip RAM becomes unreliable. At least, that's what I've heard.
And the "run it to death or replace it now" decision is complicated by the fact that the new MacBook cases are designed with zero spare space, and there is some risk of damaging other parts while attempting to replace the drive. I could pay the Apple store to do the work and have a warranty, but is a seven-year-old machine worth that expense?
Since I have both online continuous backup and local backups, I'm inclined to let it run to failure and then just replace the machine.
So thank you kindly for the advice - it HAS helped me clarify a course of action.
Boom
|
|
|
|
Post by AudioHTIT on May 28, 2020 16:03:29 GMT -5
You should also weigh the option of selling it while everything is in working condition, and getting a new one. Maybe next time someone has a MacBook sale!
|
|
KeithL
Administrator 
Posts: 9,938
|
Post by KeithL on May 28, 2020 16:06:27 GMT -5
The actual storage sites on SSDs physically wear out eventually.... and that wear is counted in "write cycles".
As I understand it, with most drives, the drive is supposed to automatically detect when faults occur and remap them. The drive also should be moving around the areas of memory being used so that the entire disc "wears evenly".
As a result, as sites wear out, you shouldn't see errors, but the capacity of the disc will start to decrease. So, in theory, it should be safe to "use it until it stops"..... (Although, when that happens, there will no longer be enough space to hold the data on it, and it will stop SUDDENLY.)
I would ALWAYS make sure that important data is backed up... and more so in this case...
Personally, I would probably "split the difference".... Keep using it for now... but keep your eye out for a good sale on a replacement. (or just replace it - considering how cheap SSDs are these days.)
|
|
|
|
Post by vcautokid on May 28, 2020 16:50:15 GMT -5
I need some help here, how does an SSD wear and how is any wear measured? In a spinner the rpm is measurable, as are the read and write speeds and the dead sectors can be counted. But an SSD, what is measured to determine wear? Cheers Gary The number of write cycles is finite. Once those figures are hit, the SSD stops writing, period. Indefinite reads, finite writes. Especially Older SATA drives have a fairly low life in direct comparison with more modern SSD drives with more advanced management and technology. |
|
|
|
Post by vcautokid on May 28, 2020 16:54:13 GMT -5
If you use SSD as a file drive, rotate your drives. I have a friend who uses an SSD Raid NAS and works peachy. SSD is the way business is done. HDD with their high error rates due to high pack density is a ticking time bomb more than an SSD. I am saving up for a NAS SSD storage. SSD drives are finally "Cheap" to buy. I saw a Samsung 500GB SSD for like a $140.00. I know I'll get some heat about HDD, but heck moving parts, gets hot, head crashes, Higher error rates than ever means more interpolation and all that stuff. Fun for the whole family you know.
|
|
|
|
Post by SteveH on May 28, 2020 16:59:04 GMT -5
I purchased two QNAP NASs and went 100% Seagate IronWolf SSD 110s in both, so I have twelve 480GB drives. I chose the SSD for less noise and less heat. The drives are warrantied for five years and the Mean Time Between Failure (MTBF) is rated at 1.8 million hours. I have had two crap out in six months and both crapped out within one week of each other. What the hell?  Seagate replaced them, but the fact that they crapped out within six months sucks and defeats the purpose of going with SSD. The same employee must have assembled that LOT after lunch on a Friday  |
|
|
|
Post by Boomzilla on May 28, 2020 19:53:07 GMT -5
If you use SSD as a file drive, rotate your drives. I have a friend who uses an SSD Raid NAS and works peachy. SSD is the way business is done. HDD with their high error rates due to high pack density is a ticking time bomb more than an SSD. I am saving up for a NAS SSD storage. SSD drives are finally "Cheap" to buy. I saw a Samsung 500GB SSD for like a $140.00. I know I'll get some heat about HDD, but heck moving parts, gets hot, head crashes, Higher error rates than ever means more interpolation and all that stuff. Fun for the whole family you know. Actually, there is a HUGE data farm here in Baton Rouge - one that I was allowed to tour. They've analyzed the reliability of SSD vs rotating drives, and found that for the $$$, they are getting far better reliability with conventional drives. They admit that this will change at some time in the future, but for now, their multiple petabytes (perhaps by now exabytes) of data reside on "old fashioned" (but "enterprise quality") spinning discs. |
|
|
|
Post by ttocs on May 28, 2020 20:31:13 GMT -5
boomzilla, have you checked your System Report? For those who are not aware, it's an easy way to see what your system thinks the drive health is. Verified is what it will say when it thinks the drive is healthy. This is how some drive utils monitor the ssd health. Here's mine:  |
|
|
|
Post by vcautokid on May 28, 2020 23:10:49 GMT -5
I purchased two QNAP NASs and went 100% Seagate IronWolf SSD 110s in both, so I have twelve 480GB drives. I chose the SSD for less noise and less heat. The drives are warrantied for five years and the Mean Time Between Failure (MTBF) is rated at 1.8 million hours. I have had two crap out in six months and both crapped out within one week of each other. What the hell? Seagate replaced them, but the fact that they crapped out within six months sucks and defeats the purpose of going with SSD. The same employee must have assembled that LOT after lunch on a Friday Seagate has had a bit of a rough on go on quality the last few years, but has been better than when I really used their products. I remember when it was Shugart. Al Shugart was running the show. From the year I graduated high school in 1978 Seagate came much later. But extend from Shugart. Ah, the days of MFM and IDE and SCSI. Yeah right, they can keep those. But fun to remember anyway. I still remember the Shugart 801 and 851 8 inch drives. You did not want to drop them on your foot. Weighed as much as a fully laidened Gaming tower computer today in some cases. Yeow! |
|
|
|
Post by millst on May 28, 2020 23:21:40 GMT -5
I don't think you'd be risking data loss, but it's worth taking care of before the drive stops working [suddenly]. Once the drive is maxed out, it will become read-only and I doubt you will be able to boot off it anymore. It will be easier to solve before that point is reached. Also, given Murphy's Law, that point will be a particularly inopportune time  |
|
|
|
Post by Boomzilla on May 29, 2020 5:35:38 GMT -5
boomzilla, have you checked your System Report? For those who are not aware, it's an easy way to see what your system thinks the drive health is. Verified is what it will say when it thinks the drive is healthy. This is how some drive utils monitor the ssd health. Here's mine: Hi ttocs - You'll note, please, that although the system report provides the identification and basic information about your HDD, it does NOT provide ANY information about the actual health of the drive. The "verified" tag only means that the drive is currently functioning. For example: It does NOT provide any information on how many sectors are bad It does NOT provide any information on the read/write cycles on the drive It does NOT provide any information on the frequency of errors It does NOT provide any information on the number of CRC errors And I could go on... In short, the System Report is totally useless in assessing the health of your drive. Always has been; always will be. The only thing that the system report COULD possibly tell you was that your drive had failed - and if that's the case, you'd know it without needing System Report. Third-party software and diagnostics are the ONLY way to get information on the health of your drive. One of the most popular utilities, "Disk Drill" has a free version that's well worth checking out. Their paid version has many more features, but most don't need it. As to reliability of specific models, my backup service, Backblaze, offers reliability stats on specific models:  They also have links to additional test results. NOTE that the Backblaze stats are garnered from data-farm drives that run continuous duty cycles. Cordially - Boom |
|
Seagate replaced them, but the fact that they crapped out within six months sucks and defeats the purpose of going with SSD. The same employee must have assembled that LOT after lunch on a Friday 
