Heavy CPU usage when accessing SATA drive

scooter41

scooter41

scooter41

Associate
Joined
11 Mar 2004
Posts
90
Hi There,

I have been victim of extreme PC lag during working over the past 3 weeks, and I couldnt work out what was causing it.

However after careful diagnostics, I have found out that 1 of my SATA drives (160gig) which is my main work drive, causing the PC to almost grind to a halt whenever the drive is either read from or written to.

I know defragging etc can speed up a drive, but this is immensely slow and I dont think its just a defrag that will solve it.

I have other SATA drives, which are all fine and I dont experience any problems from

Does anyone have any idea why this may be?

Could it be the drive is on its way out?

Any help greatley appreciated :)
 
Last edited:
scooter41 said:
Straight forward drive only, no partition!
Click to expand...
& same drive with Windows installed :eek: Anyway, that could be the cause of problem/s, because disk is sifting through a lot (files & large disk surface area) to just find a specific file. Also, what I've found, even with only a 20GB partition for Windows is, it gets fragmented quite quickly. Highly suggest you partition your disk - might solve problem but even if, highly recommeded. Acronis Disk Director or Partition Magic can create/merge/resize/etc partitions from Windows without having to reformat (recommend backing up though) :)

E.g. my setup:
mypartitions0ce.jpg


If its any issue like 'drive is on its way out', you can use a program like SpeedFan to get a Self-Monitoring, Analysis, & Reporting Technology (S.M.A.R.T.) reading of HD :D
 
Last edited:
Nope, sorry, you misunderstood me, when I said main work drive, I meant the drive with all my work on :) Not windows install....

I have 4 drives in here, doing digitial video and having large photoshop files requires lots of space.

So I really dont think fragmenting/defragmenting is the issue here.

It actually blocks up the pc so much, the mouse doesnt even move sometimes!
 
uh oh

I just run that util.......... oh dear, looks like my drives fitness is slightly low!! I think it needs to get to the gym more often!!

Seriously though, is there any cure for this or is the drive knackered?

smart.jpg
 
more analysis

I have just used the online check thing (excellent util) heres what it brought back:

smart2.jpg


Bearning in mind this has most of my work and is critical, is it suggested that I sort out a new drive pretty quick?
 
Go into Control panel>Administrative tools>Computer management>Device manager>Ide ATA/Atapi controllers, double click on the one that the drive is connected to, click on the Primary or secondary chennel tab, depending on which SATA port your drive is connected to, and make sure the transfer mode is not set to PIO. Even if its on auto, manually set ti to SATA1 or SATA2 mode. If its in PIO mode, it will cause the cpu usage to massively increase, and the transfer rate to drop drastically. It may also cause the SMART errors.
 
the primaryIDE are both set to UDMA,

however on the secondary, it looks like one is set to PIO mode... I cant change the auto detection either, could this be the problem?

smart3.jpg
 
scooter41 said:
Bearning in mind this has most of my work and is critical, is it suggested that I sort out a new drive pretty quick?
Click to expand...

If you have anything you consider "critical" - I take it you have a good backup scheme in place (whether the drive is failing or not)?!
 
If a drive errors persistantly under Windows, it usually reverts from UDMA to PIO mode. After looking at the SMART posts, it does look like the drive is knackered. One thing that may have caused it is the temp. 65*C for way too hot for a drive and I'm not surprised it's erroring out - although I don't know if the high temp is because of your setup or the fact that it's knackered and that is causing the high temps (dodgy motor, dodgy head, etc..).

Start backing up stuff NOW though. It looks like it could hold out for a while, but hard drives when they start dying, they can die pretty quickly...
 
messiah khan said:
Why would an insufficient PSU result in the harddrive using more CPU resources? That makes no sense.
Click to expand...
Was refering to the S.M.A.R.T. analysis, therefore weak/etc PSU might = damage to drive = 'using more CPU resources' :p & also he did mention 4 drives :eek: - http://www.extreme.outervision.com/psucalculator.jsp

& why the UDMA, PIO, primaryIDE, etc talk :confused: Isn't it a SATA hard disk?

Mobo & PSU specs, would really help :)
 
Last edited:
yep all backed up daily, but even so to lose a days work could be pretty annyoing ;)

I think i will try unplugging a couple of other drives, just check to see if it is a heat / power issue, then if its still knacked order a replacement
 
hp7909 said:
& why the UDMA, PIO, primaryIDE, etc talk :confused: Isn't it a SATA hard disk?
Click to expand...

Because UDMA/PIO are transfer modes for hard drives in general, not just IDE ones. Also IDE talk comes in because certain mobos/drivers/configurations mean that SATA ports show up as IDE controllers even though they aren't. It just keeps things simple.

Another recommendation is run the SMART tests on the other drives. I'd like to see what temperature the other drives are running at. If they're all running at like 60*C then there's a problem with airflow in the case and it could cause the other drives to start failing...
 
scooter41 said:
the primaryIDE are both set to UDMA,

however on the secondary, it looks like one is set to PIO mode... I cant change the auto detection either, could this be the problem?

smart3.jpg
Click to expand...


Well that is the problem that is causing the high cpu usage, and presumably slow transfer rate(It will be locked at something like 10-15mbps). Maybe have a lookin the Bios, to see if there is a setting to change the Transfer mode from auto to manual. As mentioned though, this could be a symptom of a dying drive. If the drive is 65degrees C, then it is way too hot. Have a look at the status of the other drives. Having said that though, SMART sometimes will show the temperature in fahrenheit depending on the drive make. But that would make it 18.3 degrees C which IMO would be too low for a normal reading.
 
smart4.jpg


the others are running hot too, but that could be a knock on effect of that 1 drive over heating, as they do sit pretty close together, I will try now and see what happens with just 2 drives going
 
You must log in or register to reply here.
Back
Top Bottom