PC Review


Reply
 
 
Kunal Gandhi
Guest
Posts: n/a
 
      27th Jan 2004
Hi,

I am totally new to Windows Internals and request you to bear with my
question(s).

Assumptions/Observations:

I used a utility Filemon available at www.sysinternals.com to see how the
files are accessed, and observerd that the intra-file accesses are
sequential in nature.
I observe that the application requests the data from the file in such
manner as shown by the last column where we can see the offset and the
length that is read (requested by the application which requested the read).
This length, I believe is what the application requested (and then requested
many times over to read the file).

Questions:
1. Is this length broken-up by Windows (Cache Manager?) to finer/fixed sizes
and obtained from the file?
2. If that size is fixed, what is it?
3. If not fixed, is it calculated dynamically, and how?

Pasted below, is a part of data collected using filemon.

I would appreciate any feedback on above.

Thanks and regards,

Kunal Gandhi

432 6:25:45 PM System:8 IRP_MJ_WRITE* C:\pagefile.sys SUCCESS Offset:
56483840 Length: 65536
433 6:25:45 PM System:8 IRP_MJ_WRITE* C:\pagefile.sys SUCCESS Offset:
56549376 Length: 65536
434 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 10563584 Length:
4096
435 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 25567232 Length:
16384
436 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 12288 Length: 4096
437 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 14655488 Length:
4096
438 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 15609856 Length:
4096
439 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 56442880 Length:
4096
440 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 59117568 Length:
16384
441 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 61124608 Length:
8192
442 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 62820352 Length:
4096
443 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 63012864 Length:
8192
444 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 63164416 Length:
8192
445 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 64106496 Length:
4096
446 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 28672 Length: 4096
447 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 430080 Length: 4096
448 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 462848 Length: 4096
449 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 487424 Length: 4096
450 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 2768896 Length:
4096
451 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 2850816 Length:
4096
452 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 4423680 Length:
4096
453 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 4739072 Length:
4096
454 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 6225920 Length:
4096
455 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 6967296 Length:
4096
456 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 7659520 Length:
4096


 
Reply With Quote
 
 
 
 
Pat [MSFT]
Guest
Posts: n/a
 
      27th Jan 2004
Depending on the access, Reads/Writes will need to be on page boundaries
(multiples of 4k). So, it can be dynamic but it will be a multiple.

Also, if the filesystem suspects that the file is doing sequential
reads/writes, it will attempt to optimize for that . An app can give the
filesystem hints to help this along (see CreateFile() for more information).

Pat

"Kunal Gandhi" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed)...
> Hi,
>
> I am totally new to Windows Internals and request you to bear with my
> question(s).
>
> Assumptions/Observations:
>
> I used a utility Filemon available at www.sysinternals.com to see how the
> files are accessed, and observerd that the intra-file accesses are
> sequential in nature.
> I observe that the application requests the data from the file in such
> manner as shown by the last column where we can see the offset and the
> length that is read (requested by the application which requested the
> read).
> This length, I believe is what the application requested (and then
> requested
> many times over to read the file).
>
> Questions:
> 1. Is this length broken-up by Windows (Cache Manager?) to finer/fixed
> sizes
> and obtained from the file?
> 2. If that size is fixed, what is it?
> 3. If not fixed, is it calculated dynamically, and how?
>
> Pasted below, is a part of data collected using filemon.
>
> I would appreciate any feedback on above.
>
> Thanks and regards,
>
> Kunal Gandhi
>
> 432 6:25:45 PM System:8 IRP_MJ_WRITE* C:\pagefile.sys SUCCESS Offset:
> 56483840 Length: 65536
> 433 6:25:45 PM System:8 IRP_MJ_WRITE* C:\pagefile.sys SUCCESS Offset:
> 56549376 Length: 65536
> 434 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 10563584 Length:
> 4096
> 435 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 25567232 Length:
> 16384
> 436 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 12288 Length:
> 4096
> 437 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 14655488 Length:
> 4096
> 438 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 15609856 Length:
> 4096
> 439 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 56442880 Length:
> 4096
> 440 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 59117568 Length:
> 16384
> 441 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 61124608 Length:
> 8192
> 442 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 62820352 Length:
> 4096
> 443 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 63012864 Length:
> 8192
> 444 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 63164416 Length:
> 8192
> 445 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 64106496 Length:
> 4096
> 446 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 28672 Length:
> 4096
> 447 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 430080 Length:
> 4096
> 448 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 462848 Length:
> 4096
> 449 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 487424 Length:
> 4096
> 450 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 2768896 Length:
> 4096
> 451 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 2850816 Length:
> 4096
> 452 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 4423680 Length:
> 4096
> 453 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 4739072 Length:
> 4096
> 454 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 6225920 Length:
> 4096
> 455 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 6967296 Length:
> 4096
> 456 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 7659520 Length:
> 4096
>
>



 
Reply With Quote
 
Kunal Gandhi
Guest
Posts: n/a
 
      28th Jan 2004
Thanks Pat!

Kunal Gandhi

"Pat [MSFT]" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed)...
> Depending on the access, Reads/Writes will need to be on page boundaries
> (multiples of 4k). So, it can be dynamic but it will be a multiple.
>
> Also, if the filesystem suspects that the file is doing sequential
> reads/writes, it will attempt to optimize for that . An app can give the
> filesystem hints to help this along (see CreateFile() for more

information).
>
> Pat
>
> "Kunal Gandhi" <(E-Mail Removed)> wrote in message
> news:(E-Mail Removed)...
> > Hi,
> >
> > I am totally new to Windows Internals and request you to bear with my
> > question(s).
> >
> > Assumptions/Observations:
> >
> > I used a utility Filemon available at www.sysinternals.com to see how

the
> > files are accessed, and observerd that the intra-file accesses are
> > sequential in nature.
> > I observe that the application requests the data from the file in such
> > manner as shown by the last column where we can see the offset and the
> > length that is read (requested by the application which requested the
> > read).
> > This length, I believe is what the application requested (and then
> > requested
> > many times over to read the file).
> >
> > Questions:
> > 1. Is this length broken-up by Windows (Cache Manager?) to finer/fixed
> > sizes
> > and obtained from the file?
> > 2. If that size is fixed, what is it?
> > 3. If not fixed, is it calculated dynamically, and how?
> >
> > Pasted below, is a part of data collected using filemon.
> >
> > I would appreciate any feedback on above.
> >
> > Thanks and regards,
> >
> > Kunal Gandhi
> >
> > 432 6:25:45 PM System:8 IRP_MJ_WRITE* C:\pagefile.sys SUCCESS Offset:
> > 56483840 Length: 65536
> > 433 6:25:45 PM System:8 IRP_MJ_WRITE* C:\pagefile.sys SUCCESS Offset:
> > 56549376 Length: 65536
> > 434 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 10563584

Length:
> > 4096
> > 435 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 25567232

Length:
> > 16384
> > 436 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 12288 Length:
> > 4096
> > 437 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 14655488

Length:
> > 4096
> > 438 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 15609856

Length:
> > 4096
> > 439 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 56442880

Length:
> > 4096
> > 440 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 59117568

Length:
> > 16384
> > 441 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 61124608

Length:
> > 8192
> > 442 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 62820352

Length:
> > 4096
> > 443 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 63012864

Length:
> > 8192
> > 444 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 63164416

Length:
> > 8192
> > 445 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 64106496

Length:
> > 4096
> > 446 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 28672 Length:
> > 4096
> > 447 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 430080 Length:
> > 4096
> > 448 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 462848 Length:
> > 4096
> > 449 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 487424 Length:
> > 4096
> > 450 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 2768896 Length:
> > 4096
> > 451 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 2850816 Length:
> > 4096
> > 452 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 4423680 Length:
> > 4096
> > 453 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 4739072 Length:
> > 4096
> > 454 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 6225920 Length:
> > 4096
> > 455 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 6967296 Length:
> > 4096
> > 456 6:25:45 PM System:8 IRP_MJ_WRITE* C: SUCCESS Offset: 7659520 Length:
> > 4096
> >
> >

>
>



 
Reply With Quote
 
 
 
Reply

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Java cache size moon Windows XP General 3 27th Nov 2009 06:14 AM
FBWF Cache Size archilea@gmail.com Windows XP Embedded 5 24th Mar 2007 06:43 PM
Extending the size of the temp. internet file cache, and offline webpage cache. Dave Windows XP Performance 1 29th Sep 2004 03:54 PM
cache size jebab Windows XP Performance 0 29th Dec 2003 04:01 PM
IE Browser cache Size Jimbo Windows XP General 1 21st Jul 2003 03:29 PM


Features
 

Advertising
 

Newsgroups
 


All times are GMT +1. The time now is 09:11 PM.