Re: locked page limit

Tech Tip: Click here to run a free scan for Windows Errors and optimize PC performance

From: Stanley Feng \(MSFT\) (sfeng_at_online.microsoft.com)
Date: 07/10/04


Date: Sat, 10 Jul 2004 15:44:48 -0700

Mathew,

Why don't you use a perfmon to observe NP/locked page usage?

-- 
This posting is provided "AS IS" with no warranties, and confers no rights.
Use of included script samples (if any) are subject to the terms specified
at http://www.microsoft.com/info/cpyright.htm"
"Mathew Joy" <mathewjoyatxlteknologies.com> wrote in message
news:OtYmLoUTEHA.3660@tk2msftngp13.phx.gbl...
> I've not been much clear about this for a while. Hope someone can shed
> a bit of light.
>
> Quote from MSDN
> --------------------------------------------------------------------------
--
> ---
> The limit on locked memory in Windows NT and Windows 2000 is about
> one-eighth the physical RAM for all applications combined. This is a
> rough estimate and should not be used as an exact figure on which to
> base calculations.
> --------------------------------------------------------------------------
--
> ---
>
> Though it is said that it is a rough estimte, I guess it means that it
> is almost 1/8th the total RAM. I tested this on a few machines. I
> connected one socket and issued multiple overlapped recv() calls. On
> machines (2 of them) that had 128MB of RAM, I was able to post around
> 4000 calls (which means 4000*4k = 16MB = 1/8 * 128MB). But when I
> tested it on machines with 256MB RAM I was able to post around 38000
> calls (which is almost 60% of the total RAM) and in the case of 512MB
> machines I could post around 78000 calls (which is also 60% of the
> total RAM). Considering locked pages as NP pool this is also far
> greater than the limit (1/4 of the total RAM). Could anyone explain on
> why I was able to post that many overlapped calls (all with 4K
> buffer[so assuming that one call will result only on one locked page]
> and tested on either NT or w2k server)? Mails to M$ on clarification
> was as usual mum.
>
> Thanks
>
> -- 
> Regards,
> Mathew Joy
>
>


Relevant Pages

  • Re: locked page limit
    ... > The limit on locked memory in Windows NT and Windows 2000 is about ... > rough estimate and should not be used as an exact figure on which to ... > is almost 1/8th the total RAM. ... I tested this on a few machines. ...
    (microsoft.public.win32.programmer.networks)
  • locked page limit
    ... The limit on locked memory in Windows NT and Windows 2000 is about ... rough estimate and should not be used as an exact figure on which to ... is almost 1/8th the total RAM. ... I tested this on a few machines. ...
    (microsoft.public.win32.programmer.networks)
  • locked page limit
    ... The limit on locked memory in Windows NT and Windows 2000 is about ... rough estimate and should not be used as an exact figure on which to ... is almost 1/8th the total RAM. ... I tested this on a few machines. ...
    (microsoft.public.win32.programmer.kernel)
  • Re: locked page limit
    ... Did you checked the NP pool size? ... > rough estimate and should not be used as an exact figure on which to ... > is almost 1/8th the total RAM. ... I tested this on a few machines. ...
    (microsoft.public.win32.programmer.kernel)
  • Re: locked page limit
    ... Did you checked the NP pool size? ... > rough estimate and should not be used as an exact figure on which to ... > is almost 1/8th the total RAM. ... I tested this on a few machines. ...
    (microsoft.public.win32.programmer.networks)