HDD and VMs

[javacontour]

For the second time in about as many years, the drive holding my Virtual Machine images died.

Probably just coincidence, but I was wondering if hosting a VM disk image is harder on a drive than simply hosting ordinary files like spreadsheets, documents, images and media?

I was able to recover the VM that was impacted by the bad sectors. But even f lost, it was no big deal as I use the same machines on my desktop and laptop, so I can export one from the machine that is still up, so no data lost as far as I can tell.

(See on-line back up thread as well.)

Just curious if it's just random chance the the two drives that have failed in the past two years (January 2018 and December 2019) were both the dive hosting the virtual machine disk image files, or if such work beats up the drive far more than other work loads.

 

[javacontour]

Of course, it could just be an old drive:

----------------------------------------------------------------------------
CrystalDiskInfo 5.6.2 (C) 2008-2013 hiyohiyo
Crystal Dew World : http://crystalmark.info/
----------------------------------------------------------------------------

OS : Windows 8 Pro [6.2 Build 9200] (x64)
Date : 2019/12/01 11:27:00

-- Controller Map ----------------------------------------------------------
+ Standard SATA AHCI Controller [ATA]
- hp DVDRAM GTA0N
- TOSHIBA HDWD130
- WDC WD4003FZEX-00Z4SA0
- Samsung SSD 850 PRO 512GB
- Microsoft Storage Spaces Controller [SCSI]
+ Standard NVM Express Controller [SCSI]
- MS1PC5ED3ORA3.2T

-- Disk List ---------------------------------------------------------------
(1) TOSHIBA HDWD130 : 3000.5 GB [0/0/0, pd1]
(2) WDC WD4003FZEX-00Z4SA0 : 4000.7 GB [1/0/0, pd1] - wd
(3) Samsung SSD 850 PRO 512GB : 512.1 GB [2/0/0, pd1] - sg

----------------------------------------------------------------------------
(1) TOSHIBA HDWD130
----------------------------------------------------------------------------
Model : TOSHIBA HDWD130
Firmware : MX6OACF0
Serial Number : XXXXXXXXX
Disk Size : 3000.5 GB (8.4/137.4/3000.5/3000.5)
Buffer Size : Unknown
Queue Depth : 32
# of Sectors : 5860533168
Rotation Rate : 7200 RPM
Interface : Serial ATA
Major Version : ATA8-ACS
Minor Version : ATA8-ACS version 4
Transfer Mode : SATA/600
Power On Hours : 15441 hours
Power On Count : 41 count
Temparature : 33 C (91 F)
Health Status : Caution
Features : S.M.A.R.T., APM, 48bit LBA, NCQ
APM Level : 0000h [OFF]
AAM Level : ----

=================================================

It's accumulated 643 days of powered on time.

 

[OVERKILL]

Is it harder on the drive? I'd say yes, because the drive cannot optimize the placement of anything, it's in one huge file that's just constantly thrashing the same areas.

 

[dogememe]

It's just random. I owned a hosting company with a decent amount of customers and quite a large number of physical server, most with multiple hard drives. HDD failures, while rare, seemed to happen regardless of how the drive is used. In my home I have about two dozen drives in a couple servers running 24/7... I've had one drive failure in the past year and it was not a heavily used drive.

 

[ctechbob]

https://static.googleusercontent.com/media/research.google.com/en//archive/disk_failures.pdf

Granted, this study is over 10 years old now, but its a good read for HD failures. Its a study Google did back in 2007 across their farms.

"One of our key findings has been the lack of a consistent pattern of higher failure rates for higher temperature drives or for those drives at higher utilization levels. Such correlations have been repeatedly highlighted by previous studies, but we are unable to confirm them by observing our population."


"The data in this study are collected from a large number of disk drives, deployed in several types of systems across all of Google's services. More than one hundred thousand disk drives were used for all the results presented here. The disks are a combination of serial and parallel ATA consumer-grade hard disk drives, ranging in speed from 5400 to 7200 rpm, and in size from 80 to 400 GB."

Just found that Backblaze is posting drive failure information as well.

https://www.backblaze.com/b2/hard-drive-test-data.html

 

[javacontour]

Mostly read...

But I should probably make sure I'm not swapping to the disk just in case a VM runs out of memory.

My data is stored on other drives on the host, so the VMs would be getting documents, spreadsheets, etc from the other spinning disk in the machine.

I was thinking the same with respect to the one large file, say 10 to 50gb for a disk image. As I mentioned above, if a machine begins to swap, it may get write heavy. I need to check if they have and/or are using a swap partition.

Other than the SSD, it was the oldest drive in the machine. And I do replace a number of drives with a variety of ages in my job.

It probably is harder on the drive for sure. But hard enough to cause an early death?

We can debate for ages.

Thanks for the feedback.

Originally Posted by OVERKILL
Is it harder on the drive? I'd say yes, because the drive cannot optimize the placement of anything, it's in one huge file that's just constantly thrashing the same areas.

 

[OVERKILL]

You are quite welcome