On 6/26/20 5:05 PM, Paul wrote:
philo wrote:
On 6/26/2020 4:40 PM, David W. Hodgins wrote:
On Fri, 26 Jun 2020 17:16:32 -0400, philo wrote:

I'm wondering if there are performance issues with WD- Black hard
drives.
On one of my Wife's Win10 machine, there was constant disk activity
that
I could not trace down...all seemed normal.
I decided to clone the drive to a SSD and found a nice improvement that
I thought was simply a function of the SSD.
Then realized that on my main machine running Ubuntu 20.04 that also
had
a WD-Black drive, there also seemed to be excess drive activity.
Since I
have not done a disk clone backup in years, I cloned the drive to a
Seagate SATA and found the excess drive activity had been cured.
At the time I purchased the WD-Black, I thought I was getting a top of
the line drive.

If it's a recent drive, it may be using the inferior Shingled
Magnetic Recording
(SMR) technology. See
https://www.extremetech.com/computin...drives-use-smr

for a list of which drives use the lower performance recording method.

Regards, Dave Hodgins




Thanks, but my two drives are the CMR type


At any rate, I think I will keep them just for backups

Using the available test utility from WDC, do the
short and the long test. I'm not particularly interested
in the results, just interested that, as a side-effect
of running the external utility, the internal
short and/or long testing, stops...

To find the utility, sometimes they make you hunt down
the drive model number first, then the download table
shows a test utility.

A modern hard drive *can* have sustained internal
activity, combined with a refusal to respond from the
outside. This happens with "secure erase" or "enhanced secure erase"
commands being issued. Normal software doesn't use those,
and stuff like the CMRR utility could do it. The symptoms
do not sound like a match for that one.

I'm not aware of any other patterns that can be produced
inside. There's no "wear leveling" in a hard drive. There are
little test routines that drives use, but what the output
of those would be, to the user, is a mystery. Like, if
an internal short test fails, how does the drive tell
you that, exactly ? Normally, the information would
only be imparted, if an external utility commanded it,
and the status that comes back tells you the result.

One IBM drive used to self-test, every 71 seconds,
and it was accompanied by a "screeching sound", since
the test pattern caused seeks at high speed. That might
have been on a 15K drive. It made the drive only
suitable for locked server rooms, as no human could
put up with such a noise inside a house or apartment.
But again, to what purpose ? How does the drive tell
you that something is wrong ? It can piggyback a sense
code onto some unassuming SCSI command, but with SATA
I don't know if the interface logically supports that
sort of thing.

Â*Â* Paul



Thanks Paul.


For now I have the drives put away.

Will run some tests on them next time I get a chance. Looking to the
future though, I'll be moving everything to SSD.


Seems like only yesterday I upgraded from an 850 meg drive to a 2 Gig.

I was so nervous, I broke out into a cold sweat!