Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

news.ycombinator.com

Post reply on HN
Print view View original

101–110 of 144 posts

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#101

Post by xwowsersx »

MacBook Pro (16-inch, 2021) with M1 Pro

Heavy daily development use.

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

    SMART/Health Information (NVMe Log 0x02)
    Critical Warning:                   0x00
    Temperature:                        24 Celsius
    Available Spare:                    100%
    Available Spare Threshold:          99%
    Percentage Used:                    0%
    Data Units Read:                    21,429,577 [10.9 TB]
    Data Units Written:                 22,150,889 [11.3 TB]
    Host Read Commands:                 542,140,492
    Host Write Commands:                469,241,132
    Controller Busy Time:               0
    Power Cycles:                       155
    Power On Hours:                     587
    Unsafe Shutdowns:                   10
    Media and Data Integrity Errors:    0
    Error Information Log Entries:      0

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#102

Post by krackers »

cool_hw wrote: post #7
jrib wrote: post #5

Earlier quoted context omitted.

another data point: SMART/Health Information (NVMe Log 0x02) Critical Warning: 0x00 Temperature: 30 Celsius Available Spare: 100% Available Spare Threshold: 99% Percentage Used: 1% Data Units Read: 53,345,304 [27.3 TB] Data Units Written: 59,621,038 [30.5 TB] Host Read Commands: 349,163,066 Host Write Commands: 218,226,588 Controller Busy Time: 0 Power Cycles: 109 Power On Hours: 225 Unsafe Shutdowns: 8 Media and Dat…

Your read/writes seem to be comparable to the parent, but your power on hours are significantly lower, really hard to draw any conclusions here. I wonder if this whole issue is simply MacOS dumping the stuff in memory to disk when going to standby, and then reading it all back. So for a 16GB machine, for simplicity, let's say you always have RAM saturated, and you open/close the lid 5 times in a day, you have 16GB *…

>open/close the lid 5 times in a day

It's a bit smarter than that, it only writes memory contents to disk after a given period in standby. So in your example you'd only be incurring it twice a day probably.

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#103

Post by navjack27 »

my cheapest m1 mac mini that is on all day long and is my main entertainment computer that has multiple drives connected to it... but this is the built in apple drive

  === START OF SMART DATA SECTION ===
  SMART overall-health self-assessment test result: PASSED
  
  SMART/Health Information (NVMe Log 0x02)
  Critical Warning:                   0x00
  Temperature:                        33 Celsius
  Available Spare:                    100%
  Available Spare Threshold:          99%
  Percentage Used:                    4%
  Data Units Read:                    115,425,776 [59.0 TB]
  Data Units Written:                 124,921,655 [63.9 TB]
  Host Read Commands:                 1,914,424,208
  Host Write Commands:                2,555,952,944
  Controller Busy Time:               0
  Power Cycles:                       418
  Power On Hours:                     3,500
  Unsafe Shutdowns:                   56
  Media and Data Integrity Errors:    0
  Error Information Log Entries:      0

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#104

Post by Someone »

whalesalad wrote: post #39

M1 Mac Mini. I beat the heck out of this thing. SMART/Health Information (NVMe Log 0x02) Critical Warning: 0x00 Temperature: 28 Celsius Available Spare: 100% Available Spare Threshold: 99% Percentage Used: 5% Data Units Read: 83,875,891 [42.9 TB] Data Units Written: 70,257,403 [35.9 TB] Host Read Commands: 881,926,045 Host Write Commands: 530,490,261 Controller Busy Time: 0 Power Cycles: 3,350 Power On Hours: 496 Uns…

That’s power cycles of the disk firmware, isn’t it, not power cycles of macOS? I don’t know how things work, but guess it might decide to shut down when idle.

It also could crash often, but if that’s the case, I guess you’d have seen data loss. If you run disk repair in Disk Utility, does it find problems with the SSD?

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#105

Post by WesolyKubeczek »

My numbers are nowhere near that crazy, but the first thing I always do on a Mac is disable spotlight for everything except applications and the calculator. Spotlight has always been a steaming pile of useless poo in my experience, ever since its inception.

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#106

Post by 8fingerlouie »

I have no idea how people are punishing their MacBooks that hard.

My M1 MacBook Pro (mid 2020, 16GB RAM / 512GB SSD) has been used as a daily driver for my personal stuff (not work), so while usage may be light compared to a work machine, it has seen it's fair share of C/C++/C#/Swift/Go/Python code come through (studying for a master degree), as well as papers being written.

My battery is in a much worse state than my SSD with 92% capacity remaining.

  === START OF SMART DATA SECTION ===
  SMART overall-health self-assessment test result: PASSED
  
  SMART/Health Information (NVMe Log 0x02)
  Critical Warning:                   0x00
  Temperature:                        31 Celsius
  Available Spare:                    100%
  Available Spare Threshold:          99%
  Percentage Used:                    1%
  Data Units Read:                    58,248,816 [29.8 TB]
  Data Units Written:                 33,161,373 [16.9 TB]
  Host Read Commands:                 1,190,184,018
  Host Write Commands:                543,373,537
  Controller Busy Time:               0
  Power Cycles:                       242
  Power On Hours:                     688
  Unsafe Shutdowns:                   30
  Media and Data Integrity Errors:    0
  Error Information Log Entries:      0
My "server" (iCloud content cache and backups) which is a M1 Mac mini (2020 model from 2021, 8GB/256GB), which has been running 24/7 seems to be in a "worse" state:

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

  SMART/Health Information (NVMe Log 0x02)
  Critical Warning:                   0x00
  Temperature:                        32 Celsius
  Available Spare:                    100%
  Available Spare Threshold:          99%
  Percentage Used:                    5%
  Data Units Read:                    177,894,168 [91.0 TB]
  Data Units Written:                 132,856,403 [68.0 TB]
  Host Read Commands:                 2,792,877,903
  Host Write Commands:                825,466,116
  Controller Busy Time:               0
  Power Cycles:                       239
  Power On Hours:                     1,391
  Unsafe Shutdowns:                   46
  Media and Data Integrity Errors:    0
  Error Information Log Entries:      0
Something is fishy with the power on hours though. Despite having run for over a year, it reports just under 2 months of power on hours, so it's probably not being reported correctly.

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#107

Post by pelagicAustral »

tzs wrote: post #93

Wow...looks like I don't do much disk I/O compared to most people here. I'm on a 2017 iMac that I use for both personal stuff and work. It seems like most people here can do in a few months what it took me 5 years to do. Critical Warning: 0x00 Temperature: 39 Celsius Available Spare: 100% Available Spare Threshold: 10% Percentage Used: 2% Data Units Read: 163,453,178 [83.6 TB] Data Units Written: 96,776,875 [49.5 TB]…

I'm an 2017 iMac as well, use it every day for work and personal stuff... sometimes literally hammer the stuff and my numbers look nothing like some the other ones displayed here... bit of a shocker.

   === START OF SMART DATA SECTION ===
   SMART overall-health self-assessment test result: PASSED
   
   SMART/Health Information (NVMe Log 0x02)
   Critical Warning:                   0x00
   Temperature:                        36 Celsius
   Available Spare:                    100%
   Available Spare Threshold:          10%
   Percentage Used:                    1%
   Data Units Read:                    27,459,395 [14.0 TB]
   Data Units Written:                 34,559,091 [17.6 TB]
   Host Read Commands:                 864,546,880
   Host Write Commands:                1,257,234,100
   Controller Busy Time:               3,486
   Power Cycles:                       4,595
   Power On Hours:                     1,882
   Unsafe Shutdowns:                   74
   Media and Data Integrity Errors:    0
   Error Information Log Entries:      0
   Warning  Comp. Temperature Time:    0
   Critical Comp. Temperature Time:    72858

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#108

Post by frankjr »

tzs wrote: post #93

Wow...looks like I don't do much disk I/O compared to most people here. I'm on a 2017 iMac that I use for both personal stuff and work. It seems like most people here can do in a few months what it took me 5 years to do. Critical Warning: 0x00 Temperature: 39 Celsius Available Spare: 100% Available Spare Threshold: 10% Percentage Used: 2% Data Units Read: 163,453,178 [83.6 TB] Data Units Written: 96,776,875 [49.5 TB]…

I'm surprised by the numbers here as well. I have a late 2016 MBP 512GB (used daily) and the numbers are nowhere similar. It's a bit worrying though that I'm the only one in this thread with error log entries.

  SMART/Health Information (NVMe Log 0x02)
  Critical Warning:                   0x00
  Temperature:                        30 Celsius
  Available Spare:                    100%
  Available Spare Threshold:          10%
  Percentage Used:                    3%
  Data Units Read:                    59,731,850 [30.5 TB]
  Data Units Written:                 78,972,837 [40.4 TB]
  Host Read Commands:                 1,676,492,248
  Host Write Commands:                2,466,236,096
  Controller Busy Time:               3,456
  Power Cycles:                       42,547
  Power On Hours:                     2,616
  Unsafe Shutdowns:                   131
  Media and Data Integrity Errors:    0
  Error Information Log Entries:      16

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#109

Post by petercooper »

tzs wrote: post #93

Wow...looks like I don't do much disk I/O compared to most people here. I'm on a 2017 iMac that I use for both personal stuff and work. It seems like most people here can do in a few months what it took me 5 years to do. Critical Warning: 0x00 Temperature: 39 Celsius Available Spare: 100% Available Spare Threshold: 10% Percentage Used: 2% Data Units Read: 163,453,178 [83.6 TB] Data Units Written: 96,776,875 [49.5 TB]…

I had a 16GB M1 MBP for about 9 months and I was very happy with it, but once I started using Docker, I decided I needed more RAM. Bought a 64GB one and.. wow.. it's the same CPU but the machine feels so much snappier (and it's not a clean install). macOS really seems to like having all that RAM about even if it claims to not be leaning on it heavily.

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#110

Post by cool_hw »

AdamJacobMuller wrote: post #84
cool_hw wrote: post #7

Earlier quoted context omitted.

Your read/writes seem to be comparable to the parent, but your power on hours are significantly lower, really hard to draw any conclusions here. I wonder if this whole issue is simply MacOS dumping the stuff in memory to disk when going to standby, and then reading it all back. So for a 16GB machine, for simplicity, let's say you always have RAM saturated, and you open/close the lid 5 times in a day, you have 16GB *…

Launch day purchased M1 MacBook Pro with 4TB SSD. I have custom pmset setup to NOT dump to disk (mostly) with two aliases: pmset-fast: sudo pmset -ab hibernatemode 0 sleep 0 pmset-secure: sudo pmset -ab hibernatemode 25 sleep 0 displaysleep 1 First one is memory only, if the battery dies, the battery dies and the OS goes away. The second one is hard drive only, closing the lid writes to disk only and puts the laptop…

Thank you, this is useful. I think I'm going to try the memory only option. Your Read/Writes seem lower than others (still pretty high, but much better than other Apple Silicon)
Post reply on HN