You can make a difference in the Apple Support Community!

When you sign up with your Apple Account, you can provide valuable feedback to other community members by upvoting helpful replies and User Tips.

Looks like no one’s replied in a while. To start the conversation again, simply ask a new question.

M1 Macs Excessive SSD wear and tear

I saw this in MacDaily yesterday and hoping someone might know that Apple is working on a fix.

Especially since I think these SSDs cannot be changed out. Is this a fake news story or should we be very concerned.

Posted on Feb 24, 2021 3:33 AM

Reply
Question marked as Top-ranking reply

Posted on Mar 2, 2021 3:15 AM




=== START OF SMART DATA SECTION ===

SMART overall-health self-assessment test result: PASSED


SMART/Health Information (NVMe Log 0x02)

Critical Warning: 0x00

Temperature: 35 Celsius

Available Spare: 100%

Available Spare Threshold: 99%

Percentage Used: 16%

Data Units Read: 429. 154. 771 [219 TB]

Data Units Written: 488. 723. 529 [250 TB]

Host Read Commands: 2. 057. 733. 232

Host Write Commands: 1. 228. 381. 471

Controller Busy Time: 0

Power Cycles: 124

Power On Hours: 773

Unsafe Shutdowns: 18

Media and Data Integrity Errors: 0

Error Information Log Entries: 0


My Mac mini M1 8GB 256GB has less than 3 months of life. I'm using a lot Windows 10 ARM64 Insider preview under Parallels Desktop 16.3 M1. I cannot boot from my Samsung T5 external SSD and my 256 GB SSD is soldered to the logic board. Already used 16% and writed 250 TB!!! It isn't a normal behavior. Please Apple, solve this bug ASAP or PREFERABLY recall defective M1 computers.

15 replies
Question marked as Top-ranking reply

Mar 2, 2021 3:15 AM in response to firebird0402




=== START OF SMART DATA SECTION ===

SMART overall-health self-assessment test result: PASSED


SMART/Health Information (NVMe Log 0x02)

Critical Warning: 0x00

Temperature: 35 Celsius

Available Spare: 100%

Available Spare Threshold: 99%

Percentage Used: 16%

Data Units Read: 429. 154. 771 [219 TB]

Data Units Written: 488. 723. 529 [250 TB]

Host Read Commands: 2. 057. 733. 232

Host Write Commands: 1. 228. 381. 471

Controller Busy Time: 0

Power Cycles: 124

Power On Hours: 773

Unsafe Shutdowns: 18

Media and Data Integrity Errors: 0

Error Information Log Entries: 0


My Mac mini M1 8GB 256GB has less than 3 months of life. I'm using a lot Windows 10 ARM64 Insider preview under Parallels Desktop 16.3 M1. I cannot boot from my Samsung T5 external SSD and my 256 GB SSD is soldered to the logic board. Already used 16% and writed 250 TB!!! It isn't a normal behavior. Please Apple, solve this bug ASAP or PREFERABLY recall defective M1 computers.

Mar 10, 2021 3:38 PM in response to firebird0402

How many of you experiencing this issue used the migration assistant to move your old apps, prefs, and data to your new M1 Mac?


After doing a bit of digging, I found some info about about people having excessive power drain caused by the process "kernelmanagerd" which I guess manages your extensions. It seems Big Sur doesn't like some older extensions and it was consuming CPU cycles.


For me, this same process led to over 20GB being written to my SSD in less than 1 hour after a reboot.


I have always migrated data as I was too lazy to reinstall apps, re-enter license keys, and reset all of my prefs. But after a clean install last night and reinstalling apps fresh with no data migration, the problem seems to be gone so far. As I keep reinstalling apps, I will see if the problem resurfaces, but my guess is that the issue might have been caused by 1 or more system extensions that were years (or even maybe over 1 decade) old.


Worth a try if you can back up your data.


Just remember that with a clean install of Big Sur, it's unlike previous OS's and there are additional security settings regarding boot drives and such that you should first check before wiping (otherwise reinstalling is a PAIN).

Mar 11, 2021 7:17 AM in response to AlexMacMiniM1


AlexMacMiniM1 wrote:




=== START OF SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED

SMART/Health Information (NVMe Log 0x02)
Critical Warning: 0x00
Temperature: 35 Celsius
Available Spare: 100%
Available Spare Threshold: 99%
Percentage Used: 16%
Data Units Read: 429. 154. 771 [219 TB]
Data Units Written: 488. 723. 529 [250 TB]
Host Read Commands: 2. 057. 733. 232
Host Write Commands: 1. 228. 381. 471
Controller Busy Time: 0
Power Cycles: 124
Power On Hours: 773
Unsafe Shutdowns: 18
Media and Data Integrity Errors: 0
Error Information Log Entries: 0

My Mac mini M1 8GB 256GB has less than 3 months of life. I'm using a lot Windows 10 ARM64 Insider preview under Parallels Desktop 16.3 M1. I cannot boot from my Samsung T5 external SSD and my 256 GB SSD is soldered to the logic board. Already used 16% and writed 250 TB!!! It isn't a normal behavior. Please Apple, solve this bug ASAP or PREFERABLY recall defective M1 computers.

How much RAM does your system have?

If you are using Parallels + Windows, even if it is ARM64,

you are likely using tons of swap disk all the time and is

not an Apple problem but your selection of a machine with insufficient RAM.


The super fast speeds of the M1 SSDs and the efficiency of the M1

will easily mask such insufficient RAM issues.


Personally, my experience with VMs using both Parallels and VMware Fusion

has given a rule of thumb of a bare minimum of 16 GB with 24 GB of higher

preferred for best performance.

Mar 14, 2021 8:05 AM in response to firebird0402

With this "excessive write" issue I have been doing a lot of

monitoring.


One thing that I found quite interesting. I was watching RAM consumption

on my M1 MacBook Air with a particular photo processing app. After

processing only 11 RAW images and converting them to 16 bit TIFFs,

the application consumed nearly 10 GB of RAM, which on my M1 was

definitely consuming swap and likely doing lots of R/W operations

on the SSD because of it. Now, while using, the speed and efficiency

of the M1 and the SSD made things smooth and fast. But the RAM

"bloating" had me concerned. So, with the same 11 images and exactly the

same processing I did the same on my Intel iMac with the same version of

the app. The result, on the Intel machine a the app maxed out at using only

2.3 GB of RAM.


The app is still Intel code so Rosetta was used on the M1. So, as I see it,

Rosetta is possibly not properly restoring RAM to the system when an

Intel app releases it or there is a "memory leak" in the app when run

on Rosetta that does not show itself on Intel machines.


So, after this long winded spiel, I am willing to guess that some of these

excessive writes are occurring due to some Intel apps not playing well

with Rosetta or the other way around or a mix of both. I do have some other

Intel apps that I can use to test out this idea and will see what happens.


Beyond this phenomenon, I am willing to bet that those seeing excessive writes

are people using 8GB machines with apps that need 16 GB or more and because

the speed and efficiency of the M1 architecture, they don't see the slowdowns

one would have seen in the past associate with insufficient RAM.

M1 Macs Excessive SSD wear and tear

Welcome to Apple Support Community
A forum where Apple customers help each other with their products. Get started with your Apple Account.