Slow write speeds on a WD 500GB drive

[bbbl67]

A friend of mine had an WD My Book external case, which contained a
500GB WDC WD5000AAVS-00ZTB0 hard drive inside it. At some point along
the way the My Book's interface failed, it's warranty was expired, and
so we took its hard drive out and installed it directly inside his PC
case. This was back in mid- to late-2011, that it was transferred from
external to internal. It seemed to be fine for the most part in
performance, but recently we did some benchmarking and we were shocked
to find that although its reading speeds seem normal (around 50-60 MB/
s), its write speeds were shockingly low (around 10 MB/s, if even
that). What could cause such a weird asymmetry in read vs. write
speeds? It didn't seem that slow while it was inside the My Book case.

This is one of those WD Caviar Green drives, so I'm wondering if there
is some kind of power savings setting affecting it?

Yousuf Khan

 

[Rod Speed]

A friend of mine had an WD My Book external case, which contained a
500GB WDC WD5000AAVS-00ZTB0 hard drive inside it. At some point
along the way the My Book's interface failed, it's warranty was expired,
and so we took its hard drive out and installed it directly inside his PC
case. This was back in mid- to late-2011, that it was transferred from
external to internal. It seemed to be fine for the most part in
performance,
but recently we did some benchmarking and we were shocked to find
that although its reading speeds seem normal (around 50-60 MB/
s), its write speeds were shockingly low (around 10 MB/s, if even
that). What could cause such a weird asymmetry in read vs. write
speeds? It didn't seem that slow while it was inside the My Book case.

This is one of those WD Caviar Green drives, so I'm wondering if there
is some kind of power savings setting affecting it?

Cant see why any of those would produce that effect only with writes.

They are mostly just 5400 RPM drives to get the 'green' result.

You sure its hasn't got verify after write turned on ?

Why have you changed your nick but kept the email address unmunged ?

 

[bbbl67]

Cant see why any of those would produce that effect only with writes.

They are mostly just 5400 RPM drives to get the 'green' result.

The specs seem to just try to hide the speed behind some marketing
term called IntelliSpeed or something.

You sure its hasn't got verify after write turned on ?

What is that? It sounds like something you do in software.

Why have you changed your nick but kept the email address unmunged ?

I'm posting from Google groups, that's all.

 

[Arno]

A friend of mine had an WD My Book external case, which contained a
500GB WDC WD5000AAVS-00ZTB0 hard drive inside it. At some point along
the way the My Book's interface failed, it's warranty was expired, and
so we took its hard drive out and installed it directly inside his PC
case. This was back in mid- to late-2011, that it was transferred from
external to internal. It seemed to be fine for the most part in
performance, but recently we did some benchmarking and we were shocked
to find that although its reading speeds seem normal (around 50-60 MB/
s), its write speeds were shockingly low (around 10 MB/s, if even
that). What could cause such a weird asymmetry in read vs. write
speeds? It didn't seem that slow while it was inside the My Book case.

From my experience, this sounds like the drive has problems
positioning or finding sectors. If it was finding sectrors, then
the read-speed should also be affected. My guess would be weak
power or vibration that makes the more precise head positioning on
writes difficult.

This is one of those WD Caviar Green drives, so I'm wondering if there
is some kind of power savings setting affecting it?

Not that bad.

Arno

 

[Arno]

The specs seem to just try to hide the speed behind some marketing
term called IntelliSpeed or something.

5400 rpm is only 25% less. Thoughput is usualy comparable
as the lower speed allows for higher data density.

What is that? It sounds like something you do in software.

It used to be a software setting for floppy disks in the DOS age
and it is just as irrrelevant today. It would also only about halve t
he speed, not bring it down as much as you see.

Just Rod's usual atrocious lack of insight at work.

Arno

 

[bbbl67]

From my experience, this sounds like the drive has problems
positioning or finding sectors. If it was finding sectrors, then
the read-speed should also be affected. My guess would be weak
power or vibration that makes the more precise head positioning on
writes difficult.


Not that bad.

I'm gonna switch it to the IDE driver to see if it fixes it somehow
today.

Yousuf Khan

 

[bbbl67]

I'm gonna switch it to the IDE driver to see if it fixes it somehow
today.

  Yousuf Khan

Well, it looks like that didn't work. It's still horribly slow in
writes, here's a screen cap for ATTO benchmark on it:

http://imageshack.us/photo/my-images/24/captureattowdc500gbbenc.jpg/

 

[Yousuf]

I'm gonna switch it to the IDE driver to see if it fixes it somehow
today.

Yousuf Khan

Looks like the driver change didn't help. It's still pretty bad when it comes to writes:

http://imageshack.us/photo/my-images/24/captureattowdc500gbbenc.jpg/

Yousuf Khan

 

[Arno]

Well, it looks like that didn't work. It's still horribly slow in
writes, here's a screen cap for ATTO benchmark on it:

http://imageshack.us/photo/my-images/24/captureattowdc500gbbenc.jpg/

Hmm. The write speed is pretty consistent and on the level of
512B writes. Maybe multi-sector transfer is off?

Arno

 

[Yousuf Khan]

Hmm. The write speed is pretty consistent and on the level of
512B writes. Maybe multi-sector transfer is off?

The drive also has one bad sector on it (actually uncorrectable sector),
but I can't see that causing the entire drive to become nearly unwritable.

Yousuf Khan

 

[Rod Speed]

The specs seem to just try to hide the speed behind
some marketing term called IntelliSpeed or something.

Yeah, tho the datasheets do spell that out with all the ones I have checked.

What is that?

Basically the drive reads what has just been
ritten and checks that it can read it fine.

It sounds like something you do in software.

Most drives can have it enabled.

In fact one family of drives, Maxtors from memory,
used to have it on by default with new drives and
turned it off auto when they drive had done X
power on hours or # or writes, forget which.

Basically it was someone's idea for how to wring the
maximum capacity out of a drive, be conservative
about the initial factory bad block scan and let the
drive remap new bads with that verify after write
ensuring that the users data was safe.

Dunno how many others followed that approach or if any still do.

I'm posting from Google groups, that's all.

But why cant you use your normal name with that too ?

I can when posting from groups.google.

 

[Brown Joe]

5400 rpm is only 25% less.

I clearly said that that isnt the reason for just the writes being affected.

Thoughput is usualy comparable as the
lower speed allows for higher data density.

That’s just plain wrong with those ecogreen drives.

If it was that simple they wouldn’t sell the non green drives too.

It used to be a software setting for floppy disks in the DOS age

Its actually part of the ATA standard.

and it is just as irrrelevant today.

Wrong, as always.

It would also only about halve the speed,
not bring it down as much as you see.

Only if it isnt have a problem with the writes.

Just Rod's usual atrocious lack of insight at work.

Guess which pathetic little posturing drunk has
just got egg all over its stupid little face, as always ?

--

Arno Wagner, Dr. sc. techn., Dipl. Inform., CISSP --

You should have a jpg of your dick appended...

 

[Rod Speed]

Yousuf wrote

Looks like the driver change didn't help. It's still pretty bad when it
comes to writes:

http://imageshack.us/photo/my-images/24/captureattowdc500gbbenc.jpg/

I'd try another benchmark to see if that one is getting royally confused
somehow.

 

[Yousuf Khan]

Most drives can have it enabled.

In fact one family of drives, Maxtors from memory,
used to have it on by default with new drives and
turned it off auto when they drive had done X
power on hours or # or writes, forget which.

Basically it was someone's idea for how to wring the
maximum capacity out of a drive, be conservative
about the initial factory bad block scan and let the
drive remap new bads with that verify after write
ensuring that the users data was safe.

That would actually explain why my old IDE Maxtors were so bullet proof,
even five years later. Other drives had several sectors remapped, or
even unrecoverable, after only a few months, but these Maxtors stayed
error-free after half-a-decade. Ironically I just finally got rid of
those Maxtors this week too.

But why cant you use your normal name with that too ?

I can when posting from groups.google.

I only use it when I'm posting from someone else's computer. It simply
was setup that way from before.

Yousuf Khan

 

[Yousuf Khan]

I'd try another benchmark to see if that one is getting royally confused
somehow.

It's the only benchmark that would run on that drive. CrystalMark took
forever just to write a test file to it. Even a "quick" format of the
drive was dog-slow.

Yousuf Khan

 

[Rod Speed]

Rod Speed wrote

That should have read, be LESS CONSERVATIVE

That would actually explain why my old IDE
Maxtors were so bullet proof, even five years later.

It doesn't, actually.

Other drives had several sectors remapped, or
even unrecoverable, after only a few months,

Mine never did.

but these Maxtors stayed error-free after half-a-decade.

But that approach should have seen the Maxtors
produce more remapped sectors early on than
otherwise. Sorry I rarely bother to proof read.

Ironically I just finally got rid of those Maxtors this week too.

They were actually notorious for dying
easier than most of not adequately cooled.

I only use it when I'm posting from someone else's computer.
OK.

It simply was setup that way from before.

Yeah, I can see why too given that its your yahoo.com email address.

 

[Rod Speed]

Rod Speed wrote

It's the only benchmark that would run on that drive.

CrystalMark took forever just to write a test file to it. Even a "quick"
format of the drive was dog-slow.

That does support the claim that the drive really is very slow
to write when its not just a benchmark that claims that.

I'd try the drive in another system myself in case its actually
a problem with the machine or the controller, not the drive.

 

[Arno]

The drive also has one bad sector on it (actually uncorrectable sector),
but I can't see that causing the entire drive to become nearly unwritable.

Yousuf Khan

I think the consistnt write rate points to some limiting factor
that is far more constant than a hardware problem. Unless the
disk has decided to limit its write rate due to such a problem.
Would be news to me, but sich a behaviour can certainly be
implemented.

Well, frankly I don't know. I would not trust this drive though.

Arno

 

[Franc Zabkar]

The drive also has one bad sector on it (actually uncorrectable sector),
but I can't see that causing the entire drive to become nearly unwritable.

Perhaps bad sectors are like cockroaches. When you see one, there may
be a lot more that you don't see.

In any case, AFAICS, bad sectors, or heads with weak read elements,
would show up during reads more so than writes.

- Franc Zabkar

 

[Yousuf Khan]

I think the consistnt write rate points to some limiting factor
that is far more constant than a hardware problem. Unless the
disk has decided to limit its write rate due to such a problem.
Would be news to me, but sich a behaviour can certainly be
implemented.

Well, frankly I don't know. I would not trust this drive though.

Well, either would I, so I exchanged it with a different WD 500GB drive
for him. I took his old 500GB drive and put it into an eSATA case and I
am now testing it on my own system. I'll run a full SMART test on it.

Yousuf Khan