ssd swap - high usage of Terabytes Written

[Tev11]

But does disabled the caches stop Chrome from launching all its lille "helper" processes? That according to some people is where the problem is and not with the caches themselves.

If we’re talking about the disk writes because of streaming video on YouTube, this is due to a Chrome Helper shown on Activity Monitor. If we trace the process on AM to Chrome’s Task Manager, this is specifically the Utility: Network Service, which I believe “downloads” the video on your SSD temporarily as "cache," instead of using RAM (according to an earlier post).

I’m not aware of any other helper causing the excessive disk-writes based on what I’ve noticed on my own Mac. You can trace it directly by looking at the Process Number (the PID column). Both Activity Monitor and the Chrome Task Manager display the same PID.

Disabling caching, at least to my own Mac, entails setting the Google Cache Folder and all the folders in it for Chrome to read only, and I also locked it. This way Chrome is forced to use RAM to play YouTube Videos, which from my own anecdotal experience, doesn’t even really affect RAM usage.

By doing this, the Google Chrome Helper on Activity Monitor (or Utility Network Service for Chrome Task Manager) does not write to the SSD in extraordinary amounts.

 

[Maximara]

For me heavy writes are still a thing.

M1 Air 16/256 bought on 5th of February. So 49 days so far.

smartctl 7.2 2020-12-30 r5155 [Darwin 20.3.0 arm64] (local build)
Copyright (C) 2002-20, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Number:                       APPLE SSD AP0256Q
Serial Number:                      0ba0108b61388231
Firmware Version:                   1161.80.
PCI Vendor/Subsystem ID:            0x106b
IEEE OUI Identifier:                0x000000
Controller ID:                      0
NVMe Version:                       <1.2
Number of Namespaces:               3
Local Time is:                      Mon Mar 22 21:09:38 2021 CET
Firmware Updates (0x02):            1 Slot
Optional Admin Commands (0x0004):   Frmw_DL
Optional NVM Commands (0x0004):     DS_Mngmt
Maximum Data Transfer Size:         256 Pages

Supported Power States
St Op     Max   Active     Idle   RL RT WL WT  Ent_Lat  Ex_Lat
0 +     0.00W       -        -    0  0  0  0        0       0

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

SMART/Health Information (NVMe Log 0x02)
Critical Warning:                   0x00
Temperature:                        41 Celsius
Available Spare:                    100%
Available Spare Threshold:          99%
Percentage Used:                    1%
Data Units Read:                    41 373 397 [21,1 TB]
Data Units Written:                 38 522 257 [19,7 TB]
Host Read Commands:                 250 523 492
Host Write Commands:                179 807 679
Controller Busy Time:               0
Power Cycles:                       100
Power On Hours:                     164
Unsafe Shutdowns:                   4
Media and Data Integrity Errors:    0
Error Information Log Entries:      0

Read 1 entries from Error Information Log failed: GetLogPage failed: system=0x38, sub=0x0, code=745

jarek@MacBook-Air-Jarosaw ~ %

So to make it easier to read:
21,1 TB written in 49 days. It's not fair to even it out, but let's do it. Around ~0,42 TB/day which translates to over 12,5 TB/months and 153 TB in a year. That's still very bad.

That is only part of the calculations. That 19.7 is 1%. While that could be anywhere between 0.5% to 1.499..% we'll go with the 1%. 19.7 x 100 is 1900 TBW for 100%. Even if throw that out as ridiculously high we still have 1% in 49 days which results in 100% for 4800 days or 13.42 years

No. I don't plan to stop using tabs in browsers, nor disabling Spotlight or cache. I'm using it in range of normal work use and far from any kind of super heavy corporate use (which IF happened, still isn't an excuse).

That should not happen.

By your own numbers the drive has a lifespan of 13.42 years. Even if we go by the 800 TBW that resulted from a bank doing something very very silly with an entry level machine you are still talking about ~1989 days (800/(19.7/49)/365) or 5.45 Years.

I love how good M1 performs in my normal tasks. But I as of now, I can't accept this status. If nothing changes I will be selling my MBA and switch back to Windows laptops which I also use & abuse without any kind of wear level. Yeah. I will be missing just insane battery life. But that's not enough for me if other laptop can draw power from USB-C powerbank

Please explain to us just how long do want the SSD to last if 5.45 or perhaps 13.42 years are not enough.

 

[Maximara]

Well, 19.7 TB not 21.1 but still that is too much.

Not by the standard of the percentage (1%). This the hard fact about data gathering - you can't cherry pick your data after you get it. If we can trust the amount written then why not the percentage?

We know exactly what that percentage means: "Percentage Used: This is an estimate of the amount of life used by the SSD" "However, it’s important to realize what it means to reach 100 percent of projected lifetime – it does not mean that the drive is going to fail when that counter rolls over to 101 percent, only that your SSD may need to be replaced soon".

Even we assume a total 800 TBW (based on a bank doing something really really goofy) we have ~5.4 years but since we know that 1% (could be 0.5 to 1.499...% but we will go this as the stated value) was used in 49 days then 100% will take 4900 days or ~13.4 years.

From the number in this example we have a non-problem. These are the "Cold Equations".

 

[Mr Screech]

Curious if the new 11.3 update will resolve most of this.
It added 0.5 TB of writes in a few hours on mine.

More importantly, I'd like an explanation why the SSD is writing all those TB's.
An OS that idles shouldn't write anything to a disk, besides some small log files and auto-saving of opened files.
I find it highly suspicious we see TB's of written data that seems to have no clear origin.

Also just because the drive is spec'ed to last x amount of TB's, doesn't mean the OS should be allowed to write TB's of invisible data.

 

[osplo]

An update on my M1 Macbook Pro; I'm now down to averaging SSD writes below 1GB/hour which is beyond my expectations when I set out to fix my SSD writes.

As of writing this I'm writing just 0.78GB/hour to the SSD, which is honestly even lower than I remember my intel macbook writing. On my system I am way past the point of considering this a non issue anymore - it's hilariously low now.

I am happy to know that.

Thanks a lot for sharing what worked for you. I don't have any issues with my M1 (but I am using only native apps so far), but who knows? Maybe these commands will be useful in the future.

 

[Maximara]

Curious if the new 11.3 update will resolve most of this.
It added 0.5 TB of writes in a few hours on mine.

More importantly, I'd like an explanation why the SSD is writing all those TB's.
An OS that idles shouldn't write anything to a disk, besides some small log files and auto-saving of opened files.
I find it highly suspicious we see TB's of written data that seems to have no clear origin.

Also just because the drive is spec'ed to last x amount of TB's, doesn't mean the OS should be allowed to write TB's of invisible data.

The problem is we don't know if it is the OS per say or some third party program (like Chrome) doing something silly via the kernel. The thing is if Chrome goes SSD write happy on the MacOS (which it seems to do with helpers galore) then it logically should be doing the same thing on the Windows machines with SSDs. So why no news about that?

It all goes back everyone is effectively looking in magic 8 balls, at tarot decks, and in tea leaves for answers. I have shown that these third party tools, when you follow the math to its logical conclusion, are producing nonsense.

For example, the 170TB written with smartctl reporting 2% used reported by Fomalhaut produces a totally off the wall impossible 6827 (170*100/2.49) TBW for his drive. If we go with basic math where 2% x 50 = 100% then we get 8500 TBW (170TB x 50) which is insane. I know SSD have improved since the test on how long they last was but there is just no way those numbers are real. The math shows that can't be real.

Moreover has been shown that Activity Monitory shows all writes to all drives so looking at that if you have Time Machine or iCloud connected doesn't tell you jack about what is being written to the SSD.

 

[Carlson-online]

This thread is going off on an interesting tangent now. Not using your new computer just in case it breaks??

just use it and don’t worry about this issue. Do proper time machine backups. If it blows up in the first year or so it’s under warranty anyway, so true motherboard will be replaced and you have a new shiny ssd.

life’s to short - and things like this is why we have backups and warrantys.

another similar analogy, after you not drive your car after washing it, in case it gets dirty again?

 

[leons]

Another data point, 10 day old MacBook Air, 8GB
Basically, only do web browsing.
The only "unusual" habit I have is a lot of Safari tabs open at one time, sometimes 20-40.
Almost 8TB written in 10 days, so .8TB/day.
Got number using Coconut Battery (IMHO, the easiest way to get SSD info).

At a rating of 150TBW, the drive should last me about 6 months!

 

[daverdfw]

Just updated to 11.3B5 lets see if this issue gets any better.

 

[Maximara]

Another data point, 10 day old MacBook Air, 8GB
Basically, only do web browsing.
The only "unusual" habit I have is a lot of Safari tabs open at one time, sometimes 20-40.
Almost 8TB written in 10 days, so .8TB/day.
Got number using Coconut Battery (IMHO, the easiest way to get SSD info).

At a rating of 150TBW, the drive should last me about 6 months!

View attachment 1748165

Where is that 150TBW rating coming from? 7.76/150 is 5.17% which would show as "5%" but you are showing 0%. Your numbers don't line up with a 150TBW rating. Since that "0%" could actually be as high as 0.499..., based your numbers you should have a minimum of a 1500 TBW rating ie 10 times the number you gave.

Again regarding people using these tools you need to look at all the data not just a subsection of it.

 

[leons]

Admittedly, I'm not very knowledgeable on these numbers. But, the 150TBW rating came from a recent ZDNet news article: "Typically, a 250GB SSD will have a rating of 60 and 150TBW". Not even sure what the "60" refers to.

 

[Kung gu]

Another data point, 10 day old MacBook Air, 8GB
Basically, only do web browsing.
The only "unusual" habit I have is a lot of Safari tabs open at one time, sometimes 20-40.
Almost 8TB written in 10 days, so .8TB/day.
Got number using Coconut Battery (IMHO, the easiest way to get SSD info).

At a rating of 150TBW, the drive should last me about 6 months!

View attachment 1748165

This does not even make sense. 7.76TB written but lifetime used is 0%. Yeah I definitely think that those data readings are wrong. I would not trust coconut batter reading at all

 

[Kung gu]

Admittedly, I'm not very knowledgeable on these numbers. But, the 150TBW rating came from a recent ZDNet news article: "Typically, a 250GB SSD will have a rating of 60 and 150TBW". Not even sure what the "60" refers to.

its means a 256 ssd will have a rating ranging from 60TBW to 150TBW. and higher capacity drives will have higher TBW

 

[jdb8167]

Not by the standard of the percentage (1%). This the hard fact about data gathering - you can't cherry pick your data after you get it. If we can trust the amount written then why not the percentage?

We know exactly what that percentage means: "Percentage Used: This is an estimate of the amount of life used by the SSD" "However, it’s important to realize what it means to reach 100 percent of projected lifetime – it does not mean that the drive is going to fail when that counter rolls over to 101 percent, only that your SSD may need to be replaced soon".

Even we assume a total 800 TBW (based on a bank doing something really really goofy) we have ~5.4 years but since we know that 1% (could be 0.5 to 1.499...% but we will go this as the stated value) was used in 49 days then 100% will take 4900 days or ~13.4 years.

From the number in this example we have a non-problem. These are the "Cold Equations".

The TBW value is measured but the Percentage Used is estimated by the vendor, in this case Apple. Measurements can be tested by comparing known write amounts with the value returned. I've done this and so have others. The measured amount of data written appears to be correct. On the other hand, if Apple made a mistake in any of these values, the most likely place would be in an estimate. Don't you think?

 

[abhi182]

Just to add some more data
3 months , 6.59TBW
0% used

Usage is limited to mostly native programs with the notable exception of MS teams .
Majority of usage is productivity apps/ Zoom/ Teams and of course, browsing (99% Safari)

On an average, I see around 70-90GB written every day I use the system for a full day .. and 20-40GB on days of intermittent usage

Now based on the above it would appear at first glance that the endurance of the drive is at least 660TB or more
And thus I should have no reason to worry about it as it would be at least 50*3 = 150 months before the drive crosses 50% use (50% being an arbitrary threshold beyond which I may need to look at disposal/ replacement(

Yet I keep looking at these stats because I am not sure if the usage stats are linear - or if they start rising only after the spare/ redundant blocks on the SSD are used up .
In other words, if the latter is true then the change from 0 to 1% will take a long time but will rise a lot more rapidly after

 

[Maximara]

Admittedly, I'm not very knowledgeable on these numbers. But, the 150TBW rating came from a recent ZDNet news article: "Typically, a 250GB SSD will have a rating of 60 and 150TBW". Not even sure what the "60" refers to.

"Also, manufacturers are now offering higher levels of TBW -- like Samsung's 600TBW rating backed by a 10-year warranty -- so this is rapidly becoming a non-issue." also appears in the article.

 

[leons]

"Also, manufacturers are now offering higher levels of TBW -- like Samsung's 600TBW rating backed by a 10-year warranty -- so this is rapidly becoming a non-issue." also appears in the article.

Not sure I would assume that Apple is putting high-end SSDs in $999 machines. Interesting that they haven't made a peep about this.

 

[Maximara]

The TBW value is measured but the Percentage Used is estimated by the vendor, in this case Apple. Measurements can be tested by comparing known write amounts with the value returned. I've done this and so have others. The measured amount of data written appears to be correct. On the other hand, if Apple made a mistake in any of these values, the most likely place would be in an estimate. Don't you think?

I have pointed out before, referring to Apple M1 SSD Lifespan Ageing. Do YOU have the problem?, that Activity Monitor is only useful as a crosscheck if no other drives are connected to the computer because it shows all writes to all drives.

People using iCloud and Time Machine are among those who have said Activity Monitor backs up what these tools are telling them...which if correct then something is clearly wrong with the tools because those writes going to the SSD and either iCloud or the Time Machine drive. They are not all going to the SSD and yet supposedly that is what the tools being used say.

 

[wirtandi]

Curious if the new 11.3 update will resolve most of this.
It added 0.5 TB of writes in a few hours on mine.

More importantly, I'd like an explanation why the SSD is writing all those TB's.
An OS that idles shouldn't write anything to a disk, besides some small log files and auto-saving of opened files.
I find it highly suspicious we see TB's of written data that seems to have no clear origin.

Also just because the drive is spec'ed to last x amount of TB's, doesn't mean the OS should be allowed to write TB's of invisible data.

Yes, I have been noticing a trend , where 11.3 finally fixes it

 

[Maximara]

Not sure I would assume that Apple is putting high-end SSDs in $999 machines. Interesting that they haven't made a peep about this.

I have explained several times why they would keep quiet...there are just too many variabilities to make anything resembling an announcement. As you stated Apple is the vender and since they don't have to pay somebody else they can put a higher grade SSD in a $999 machine. More over Samsung's high end SSDs clock in at 2400 TBW.

 

[jdb8167]

I have pointed out before, referring to Apple M1 SSD Lifespan Ageing. Do YOU have the problem?, that Activity Monitor is only useful as a crosscheck if no other drives are connected to the computer because it shows all writes to all drives.

People using iCloud and Time Machine are among those who have said Activity Monitor backs up what these tools are telling them...which if correct then something is clearly wrong with the tools because those writes going to the SSD and either iCloud or the Time Machine drive. They are not all going to the SSD and yet supposedly that is what the tools being used say.

I tested against a single drive. I used a known amount of data to write. I only had the internal drive mounted. Like I said, it is easy to test. You are making up problems where none exist.

 

[Maximara]

I tested against a single drive. I used a known amount of data to write. I only had the internal drive mounted. Like I said, it is easy to test. You making up problems where none exist.

Looking around for more information I found Smartmontool Reporting abnormally high SSD write activities? and something caught my attention. Somebody ran these tools on a 4-year old MBP, Percentage Used: 100%, Available Spare Threshold: 10%; 448 TBW. Sadly he didn't state the size of the drive but it gives us something more to work with.

 

[leons]

I have explained several times why they would keep quiet...there are just too many variabilities to make anything resembling an announcement. As you stated Apple is the vender and since they don't have to pay somebody else they can put a higher grade SSD in a $999 machine. More over Samsung's high end SSDs clock in at 2400 TBW.

However, watching the informative video that you supplied, Apple M1 SSD Lifespan Ageing. Do YOU have the problem?, my 256GB SSD in my Air has a DWPD (daily writes per day) of .3x256GB=76.8GBx10 days I have had my machine= 768GB allowed per day to stay in spec/warranty according to your expert. However, I am up to almost 8TB for that period, a multiple of over 10 times the "acceptable" amount that he mentions. Even allowing for some miscounted cloud data as you suggest (I do very little in the cloud), the number is "off the charts" too much.

 

[Fomalhaut]

The problem is we don't know if it is the OS per say or some third party program (like Chrome) doing something silly via the kernel. The thing is if Chrome goes SSD write happy on the MacOS (which it seems to do with helpers galore) then it logically should be doing the same thing on the Windows machines with SSDs. So why no news about that?

It all goes back everyone is effectively looking in magic 8 balls, at tarot decks, and in tea leaves for answers. I have shown that these third party tools, when you follow the math to its logical conclusion, are producing nonsense.

For example, the 170TB written with smartctl reporting 2% used reported by Fomalhaut produces a totally off the wall impossible 6827 (170*100/2.49) TBW for his drive. If we go with basic math where 2% x 50 = 100% then we get 8500 TBW (170TB x 50) which is insane. I know SSD have improved since the test on how long they last was but there is just no way those numbers are real. The math shows that can't be real.

Moreover has been shown that Activity Monitory shows all writes to all drives so looking at that if you have Time Machine or iCloud connected doesn't tell you jack about what is being written to the SSD.

And this uncertainty about the validity of the smartctl numbers is part of the problem.

My MBP16 with its 172TB written @2% used according to smartctl gives an unlikely results for the maximum TBW. We can't discount that it may be wrong and that the percentage is being incorrectly calculated. Or if it means something other than described in the documentation.

We can't necessarily trust any of the estimates for the estimated duration of the SSD if there are experimental observations that cast doubt on the numbers.

 

[Fomalhaut]

Admittedly, I'm not very knowledgeable on these numbers. But, the 150TBW rating came from a recent ZDNet news article: "Typically, a 250GB SSD will have a rating of 60 and 150TBW". Not even sure what the "60" refers to.

Re-read the quote and my answer and thought it was a bit snarky! I "interpreted" the "between 60 and 150TBW", but it doesn't say that. It says a rating of 60 (what?) and 150TB - which could imply the "60" referred to some other metric. My bad!

It's obviously 60 TBW.... the same way "between 50 and 70 years old" is unambiguously referring to the range of the same unit (years).