From hijiangwen at gmail.com  Wed Apr  9 10:21:08 2014
From: hijiangwen at gmail.com (Evan Jiang)
Date: Wed, 9 Apr 2014 22:21:08 +0800
Subject: [Disksim-users] Questions about modifying disk cache
Message-ID: <CACerO_s87SM+MQi1W14t2PYQgzhucJTf5QdEJd6B98GLwSSdxA@mail.gmail.com>

 Hi!
I want to add a algorithm to the disk cache module on controller. As the
comments of the disksim_diskcache.c is not enough to new developer, so can
anyone give some key position I need to notice ?
Thanks!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20140409/953e157c/attachment.html>

From hijiangwen at gmail.com  Wed Apr  9 22:04:20 2014
From: hijiangwen at gmail.com (Evan Jiang)
Date: Thu, 10 Apr 2014 10:04:20 +0800
Subject: [Disksim-users] About extraq and intq
Message-ID: <CACerO_vMv6dGBH4Cp3nkq5G3ZX+cRr80tHKnXM-r5R=ydr-+1g@mail.gmail.com>

Hello !
Can anyone explain  (1) what are  the extraq and intq used for ?  (2) and
when are they initialized ?  (3) what case in which new members are added
to the two queue and when are the members removed.
Thanks very much!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20140410/c5703db2/attachment.html>

From rsalarifardce87 at gmail.com  Thu Apr 10 15:51:07 2014
From: rsalarifardce87 at gmail.com (Raziye Salarifard)
Date: Thu, 10 Apr 2014 12:51:07 -0700
Subject: [Disksim-users] About extraq and intq (Evan Jiang)
Message-ID: <CAK=TG576XB=fmgfXGwnhNo_hZBkUszg34_d3hPhJmaxyLzgfFg@mail.gmail.com>

Hi

1- extq is queue of events that disksim responded to them, intq is queue of
events that disksim will respond to them!
2- may be when disksim initialized!
3- when disksim needs create a new event it first checks extq and uses
members of this queue as a new event then remove a member of it and adds
this new event to intq, later when disksim responds to an event of intq
moves this event to extq!

regards


On Thu, Apr 10, 2014 at 9:00 AM, <disksim-users-request at ece.cmu.edu> wrote:

> Send Disksim-users mailing list submissions to
>         disksim-users at ece.cmu.edu
>
> To subscribe or unsubscribe via the World Wide Web, visit
>         https://sos.ece.cmu.edu/mailman/listinfo/disksim-users
> or, via email, send a message with subject or body 'help' to
>         disksim-users-request at ece.cmu.edu
>
> You can reach the person managing the list at
>         disksim-users-owner at ece.cmu.edu
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Disksim-users digest..."
>
>
> Today's Topics:
>
>    1. About extraq and intq (Evan Jiang)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Thu, 10 Apr 2014 10:04:20 +0800
> From: Evan Jiang <hijiangwen at gmail.com>
> Subject: [Disksim-users] About extraq and intq
> To: disksim-users at ece.cmu.edu
> Message-ID:
>         <CACerO_vMv6dGBH4Cp3nkq5G3ZX+cRr80tHKnXM-r5R=
> ydr-+1g at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hello !
> Can anyone explain  (1) what are  the extraq and intq used for ?  (2) and
> when are they initialized ?  (3) what case in which new members are added
> to the two queue and when are the members removed.
> Thanks very much!
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://sos.ece.cmu.edu/pipermail/disksim-users/attachments/20140410/c5703db2/attachment-0001.html
> >
>
> ------------------------------
>
> _______________________________________________
> Disksim-users mailing list
> Disksim-users at ece.cmu.edu
> https://sos.ece.cmu.edu/mailman/listinfo/disksim-users
>
>
> End of Disksim-users Digest, Vol 98, Issue 2
> ********************************************
>



-- 
*Raziye Salarifard*
*M.Sc. Student*
*Data Storage Systems and Network Laboratory,*

*Computer Engineering Department,Sharif University of Technology*

*URL: ce.sharif.edu/~salarifard <http://ce.sharif.edu/~salarifard> *
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20140410/eb5ae895/attachment.html>

From hijiangwen at gmail.com  Tue Apr 15 03:31:33 2014
From: hijiangwen at gmail.com (Evan Jiang)
Date: Tue, 15 Apr 2014 15:31:33 +0800
Subject: [Disksim-users] Ask for the explanation of types of RAM on disk
	drive controllers
Message-ID: <CACerO_tV8b0qAqp-zm2GcaeMzUDeLqmewJQyWeEPjP=Y3aMk5Q@mail.gmail.com>

I have read a document which says the following 4 things are done by the
RAM embedded on disk driver controller. But I don't know what's difference
between buffer and cache. Thanks!

RAM on disk drive controllers
1 firmware
2 speed matching buffer
3 prefetching buffer
4 cache
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20140415/005b4787/attachment.html>

From hijiangwen at gmail.com  Sun Apr 20 09:17:36 2014
From: hijiangwen at gmail.com (Evan Jiang)
Date: Sun, 20 Apr 2014 21:17:36 +0800
Subject: [Disksim-users] About the validate file
Message-ID: <CACerO_t88Pe9Kiyj_gdFh7mLyZEMnnPLZ0mWC2WszyO0D+5i8w@mail.gmail.com>

Hi, everyone!   I am reading the disk buffer module of disksim v1.0.
1 Can anyone explain what are the  "Doub","Trip","Miss" and "Hit" used for
?  Do they indicate the what the simulator should do ? for example,
simulator will adopt miss strategy when the simulator reads  "Miss" .  Or
are they used for checking results ?   For example ,  simulator use the
"Miss" to check the result . The following code just do the thing , but I
don't think the "validatebuf" will affect the buffer strategy while
searching source code.

   if (strcmp(validate_buffaction, "Doub") == 0) {
      validatebuf[0]++;
   } else if (strcmp(validate_buffaction, "Trip") == 0) {
      validatebuf[1]++;
   } else if (strcmp(validate_buffaction, "Miss") == 0) {
      validatebuf[2]++;
   } else if (strcmp(validate_buffaction, "Hit") == 0) {
      validatebuf[3]++;


2 I changed "Number of buffer segments" from 2 to 32 , but from output
file, "System logorg #0 disk Buffer hit ratio" just changes from 1561 to
1568 . It's amazing . Why can't I get a big hit increase of buffer hit ?

Thanks!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20140420/bdda4582/attachment.html>

From hijiangwen at gmail.com  Sun Apr 27 08:30:12 2014
From: hijiangwen at gmail.com (Evan Jiang)
Date: Sun, 27 Apr 2014 20:30:12 +0800
Subject: [Disksim-users] Hello, some questions about disksim buffer. Thanks .
Message-ID: <CACerO_tuJhZjmE=1cHHvoHe-zCA+D3DPVdi-QpvHJj1oywnQzQ@mail.gmail.com>

I tried the disksim v4.0. The content of my ascii trace file is as the
following:
13.191135  0   7777    5   1
 14.191135  0   7777    5   1
 15.191135  0   7777    5   1
 16.191135  0   7777    5   1
 17.191135  0   7777    5   1

All the requests is to read the blocks of the same position on disk. So the
number of buffer read hits should be 4, except the first one. However the
number of buffer accesses is 1 and hit ration is 0. The result is confusing.
Can anyone explain ? Thanks, I am looking forward to your reply. The DISK
STATISTICS is as the following.

Disk Number of buffer accesses:    1
Disk Buffer hit ratio:             0 0.000000
Disk Buffer miss ratio:                 1 1.000000
Disk Buffer read hit ratio:             0 0.000000 0.000000
Disk Buffer prepend hit ratio:            0 0.000000
Disk Buffer append hit ratio:            0 0.000000
Disk Write combinations:                0 0.000000
Disk Ongoing read-ahead hit ratio:      0 0.000000 0.000000
Disk Average read-ahead hit size:  0.000000
Disk Average remaining read-ahead: 0.000000
Disk Partial read hit ratio:      0 0.000000 0.000000
Disk Average partial hit size:     0.000000
Disk Average remaining partial:    0.000000
Disk Total disk bus wait time: 0.000000
Disk Number of disk bus waits: 0
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20140427/21dfc9c6/attachment.html>