Vista - Readyboost service and high CPU utilisation

Hi,

For the past couple of days, I've noticed my CPU usage is very high at idle
(>50% on one core of my Athlon64x2).

I finally got annoyed enough to try to work out what it was, and it turns
out it was the readyboost service (I used task manager and the process
explorer). This is odd, because I don't have any USB drives attached to the
system. What's stranger is that it happened at every boot, and I was unable
to stop the service while it was running.

I've now disabled the service at startup, and we're back to normal CPU
utilisation levels.

Does anyone have an idea as to what could be causing this?

Thanks,

Edwin

There is a ReadyBoost cache on the HD whether or not you have an external
device and is used primarily for non-sequential read/write to the HD.

Turning off the service reduces CPU usage (mostly during "idle" time),
however depending upon what activity is going on performance will probably
suffer.

"E-" <E-@discussions.microsoft.com> wrote in message
news:0CED13CD-4036-4F54-929F-CE3F2CAFC388@microsoft.com...
> Hi,
>
> For the past couple of days, I've noticed my CPU usage is very high at
> idle
> (>50% on one core of my Athlon64x2).
>
> I finally got annoyed enough to try to work out what it was, and it turns
> out it was the readyboost service (I used task manager and the process
> explorer). This is odd, because I don't have any USB drives attached to
> the
> system. What's stranger is that it happened at every boot, and I was
> unable
> to stop the service while it was running.
>
> I've now disabled the service at startup, and we're back to normal CPU
> utilisation levels.
>
> Does anyone have an idea as to what could be causing this?
>
> Thanks,
>
> Edwin

Thanks - any idea what could be causing this?

My system has a (possibly related) glitch that from time to time, it hangs
and waits for the HDD (light on) for 10-15 seconds. Although a bit of an
annoyance, I've discovered that if this happens while media centre is
recording, it'll cause a bluescreen I've been chalking it up to the file
copy problem, though I'm not sure this is the case.

The system is an Asus MNPV-VM (nVidia 430 chipset with integrated 6150
graphics) with a Samsung spinpoint P120 250Gb HDD running Home Premium x64.

Thanks again,

Edwin

"AJR" wrote:

> There is a ReadyBoost cache on the HD whether or not you have an external
> device and is used primarily for non-sequential read/write to the HD.
>
> Turning off the service reduces CPU usage (mostly during "idle" time),
> however depending upon what activity is going on performance will probably
> suffer.
>
> "E-" <E-@discussions.microsoft.com> wrote in message
> news:0CED13CD-4036-4F54-929F-CE3F2CAFC388@microsoft.com...
> > Hi,
> >
> > For the past couple of days, I've noticed my CPU usage is very high at
> > idle
> > (>50% on one core of my Athlon64x2).
> >
> > I finally got annoyed enough to try to work out what it was, and it turns
> > out it was the readyboost service (I used task manager and the process
> > explorer). This is odd, because I don't have any USB drives attached to
> > the
> > system. What's stranger is that it happened at every boot, and I was
> > unable
> > to stop the service while it was running.
> >
> > I've now disabled the service at startup, and we're back to normal CPU
> > utilisation levels.
> >
> > Does anyone have an idea as to what could be causing this?
> >
> > Thanks,
> >
> > Edwin
>
>
>

One additional bit of information (in case it's useful) - Process explorer
finds the following thread using 50% CPU time (1 full core!) on one of the
service host processes:

emdmgmt.dll!EMDMgmtGetCacheStats+0x673c

"AJR" wrote:

> There is a ReadyBoost cache on the HD whether or not you have an external
> device and is used primarily for non-sequential read/write to the HD.
>
> Turning off the service reduces CPU usage (mostly during "idle" time),
> however depending upon what activity is going on performance will probably
> suffer.
>
> "E-" <E-@discussions.microsoft.com> wrote in message
> news:0CED13CD-4036-4F54-929F-CE3F2CAFC388@microsoft.com...
> > Hi,
> >
> > For the past couple of days, I've noticed my CPU usage is very high at
> > idle
> > (>50% on one core of my Athlon64x2).
> >
> > I finally got annoyed enough to try to work out what it was, and it turns
> > out it was the readyboost service (I used task manager and the process
> > explorer). This is odd, because I don't have any USB drives attached to
> > the
> > system. What's stranger is that it happened at every boot, and I was
> > unable
> > to stop the service while it was running.
> >
> > I've now disabled the service at startup, and we're back to normal CPU
> > utilisation levels.
> >
> > Does anyone have an idea as to what could be causing this?
> >
> > Thanks,
> >
> > Edwin
>
>
>

Hi,

Just on the off chance people look for this problem via Google (Vista hangs
or freezes from time to time) I think I've cracked both problems I've
described.

The readyboost problem seems to have been a hard disk error - I ran chkdsk
/r and it identified an error in one of the readboost files, and since then
it's been fine.

The problem with Vista hanging for 20-30 seconds every once in a while is
much more widely known - googling it gets loads of results, but I'm fairly
confident I've cracked it (on my system in any case).

The problem seems to be with the nVidia nForce driver (or controller) - when
I finally started digging through my event log, I found loads of these:
Log Name: System
Source: nvstor64
Event ID: 129
Task Category: None
Level: Warning
Reset to device, \Device\RaidPort0, was issued.
Event Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="nvstor64" />
<EventID Qualifiers="32772">129</EventID>
<Level>3</Level>
<Task>0</Task>
<Keywords>0x80000000000000</Keywords>
<TimeCreated SystemTime="2007-05-28T18:49:29.084Z" />
<EventRecordID>26708</EventRecordID>
<Channel>System</Channel>
<Security />
</System>
<EventData>
<Data>\Device\RaidPort0</Data>

<Binary>0F001800010000000000000081000480040000000000000000000000000000000000000000000000000000000000000000000000810004800000000000000000</Binary>
</EventData>
</Event>

Googling turned up (only!) one hit on this
(http://www.evga.com/community/messag...TOPIC_ID=30255) but
their solution works a treat:
Disable Native Command Queueing on the SATA drive, and the problem is
solved. The explanation that disabling NCQ also improves performance strikes
me as not necessarily correct, but I haven't noticed any performance issues,
and I've now recorded Ocean's 11 off my TV card without a single bluescreen
(whereas before I'd get them every few minutes).

I hope this is helpful to others - it's definitely improved my computing
experience!

Regards,

Edwin

"E-" wrote:

> One additional bit of information (in case it's useful) - Process explorer
> finds the following thread using 50% CPU time (1 full core!) on one of the
> service host processes:
>
> emdmgmt.dll!EMDMgmtGetCacheStats+0x673c
>
> "AJR" wrote:
>
> > There is a ReadyBoost cache on the HD whether or not you have an external
> > device and is used primarily for non-sequential read/write to the HD.
> >
> > Turning off the service reduces CPU usage (mostly during "idle" time),
> > however depending upon what activity is going on performance will probably
> > suffer.
> >
> > "E-" <E-@discussions.microsoft.com> wrote in message
> > news:0CED13CD-4036-4F54-929F-CE3F2CAFC388@microsoft.com...
> > > Hi,
> > >
> > > For the past couple of days, I've noticed my CPU usage is very high at
> > > idle
> > > (>50% on one core of my Athlon64x2).
> > >
> > > I finally got annoyed enough to try to work out what it was, and it turns
> > > out it was the readyboost service (I used task manager and the process
> > > explorer). This is odd, because I don't have any USB drives attached to
> > > the
> > > system. What's stranger is that it happened at every boot, and I was
> > > unable
> > > to stop the service while it was running.
> > >
> > > I've now disabled the service at startup, and we're back to normal CPU
> > > utilisation levels.
> > >
> > > Does anyone have an idea as to what could be causing this?
> > >
> > > Thanks,
> > >
> > > Edwin
> >
> >
> >