From shanicemsuya at gmail.com  Mon May  2 16:12:19 2011
From: shanicemsuya at gmail.com (shanice msuya)
Date: Tue, 3 May 2011 05:12:19 +0900
Subject: [Disksim-users] Help with Segmentation Fault
Message-ID: <BANLkTinBw8g7vpLAykHFe1n2Aw_iJCNQMg@mail.gmail.com>

Hello DiskSim users,
I was trying to configure open synthetic workload, using one disk
Seagate_Cheetah15k disk. From the disksim 4.0 manual, it is said that
configuration can be done by setting the probability of  critical time to 0,
the probability of limited time to 0, and setting number of processor to be
equal to number of generators at 'system-level'.
I tried to set the probabilities to zero, however when I run, it gives me
'segmentation fault'.

But also I didn't understand 'system-level', where is the 'system-level'
that I should change, I assume its the parameter file, however after I
change that also its still gives me 'segmentation fault'


Shanice.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110503/c6ab69a3/attachment.html>

From chensinhome at gmail.com  Thu May  5 18:56:23 2011
From: chensinhome at gmail.com (=?Big5?B?q0inu7Ov?=)
Date: Fri, 6 May 2011 06:56:23 +0800
Subject: [Disksim-users] Disksim-users Digest, Vol 66, Issue 7
In-Reply-To: <mailman.1.1302537606.16849.disksim-users@ece.cmu.edu>
References: <mailman.1.1302537606.16849.disksim-users@ece.cmu.edu>
Message-ID: <BANLkTim08Cnr80Rkt+TaVyLKoFcFuz2Knw@mail.gmail.com>

2011/4/12 <disksim-users-request at ece.cmu.edu>

> 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. number of parallel units (f.rajaei)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Sun, 10 Apr 2011 20:40:26 -0700
> From: "f.rajaei" <frajaei55 at gmail.com>
> Subject: [Disksim-users] number of parallel units
> To: disksim-users at ece.cmu.edu
> Message-ID: <BANLkTikH1B6u2hTvyVaTP4XO7QNv6i2QDg at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hi All,
>
> Does anybody know that number of parallel unit in a parameter file for SSD
> extension is related to planes or elements?
>

   neither planes and elements. it depend on your setting

   according to parameter file
   --------
     # how many parallel units are there?
     # entire elem = 1, two dies = 2, four plane-pairs = 4
     Number of parallel units = 8, "
 ----------------------------------------------------

       e.g.:
      "entire elem " means   that  all planes in package can not work
independently















>
> Regards,
> Farzaneh
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://sos.ece.cmu.edu/pipermail/disksim-users/attachments/20110410/bda4a8b9/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 66, Issue 7
> ********************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110506/e24e06a0/attachment.html>

From shanicemsuya at gmail.com  Sat May  7 11:03:29 2011
From: shanicemsuya at gmail.com (shanice msuya)
Date: Sun, 8 May 2011 00:03:29 +0900
Subject: [Disksim-users] Help with Segmentation Fault
In-Reply-To: <BANLkTinBw8g7vpLAykHFe1n2Aw_iJCNQMg@mail.gmail.com>
References: <BANLkTinBw8g7vpLAykHFe1n2Aw_iJCNQMg@mail.gmail.com>
Message-ID: <BANLkTimNa-g8n2vS30MzJ1V0_BbvXh_XOw@mail.gmail.com>

Hello,
Please somebody help me with known tools which can generate block I/O
synthetic workloads for open and closed models.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110508/0b4e49be/attachment.html>

From shanicemsuya at gmail.com  Sat May  7 11:04:47 2011
From: shanicemsuya at gmail.com (shanice msuya)
Date: Sun, 8 May 2011 00:04:47 +0900
Subject: [Disksim-users] block I/O synthetic workload tools
Message-ID: <BANLkTinDga-GUirjNaEOTKqDmYRTmSxTmQ@mail.gmail.com>

Hello,
Please somebody help me with known tools which can generate block I/O
synthetic workloads for open and closed models.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110508/e7c92a44/attachment.html>

From haghdoost at gmail.com  Sat May  7 11:09:29 2011
From: haghdoost at gmail.com (Alireza Haghdoost)
Date: Sat, 7 May 2011 19:39:29 +0430
Subject: [Disksim-users] block I/O synthetic workload tools
In-Reply-To: <BANLkTinDga-GUirjNaEOTKqDmYRTmSxTmQ@mail.gmail.com>
References: <BANLkTinDga-GUirjNaEOTKqDmYRTmSxTmQ@mail.gmail.com>
Message-ID: <BANLkTik-S2h7ee_x9Vwg7WP8cDe97E7Pig@mail.gmail.com>

postmark, iozone, intel open storage toolkit , vdbench ...


Alireza Haghdoost
Graduate Student
Department of Computer Engineering
Sharif University of Technology
http://ce.sharif.edu/~haghdoost
Phone: (98) (21) 6616 6689




2011/5/7 shanice msuya <shanicemsuya at gmail.com>

> Hello,
> Please somebody help me with known tools which can generate block I/O
> synthetic workloads for open and closed models.
> _______________________________________________
> Disksim-users mailing list
> Disksim-users at ece.cmu.edu
> https://sos.ece.cmu.edu/mailman/listinfo/disksim-users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110507/4e9a8a4e/attachment.html>

From chensinhome at gmail.com  Sat May  7 18:28:13 2011
From: chensinhome at gmail.com (=?Big5?B?q0inu7Ov?=)
Date: Sun, 8 May 2011 06:28:13 +0800
Subject: [Disksim-users] Disksim-users Digest, Vol 67, Issue 3
In-Reply-To: <mailman.1.1304784004.28292.disksim-users@ece.cmu.edu>
References: <mailman.1.1304784004.28292.disksim-users@ece.cmu.edu>
Message-ID: <BANLkTi=tGfBKhhtpvYDtBsMkSg6detsx6A@mail.gmail.com>

"libaio"  ?


2011/5/8 <disksim-users-request at ece.cmu.edu>

> 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. Re: Help with Segmentation Fault (shanice msuya)
>   2. block I/O synthetic workload tools (shanice msuya)
>   3. Re: block I/O synthetic workload tools (Alireza Haghdoost)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Sun, 8 May 2011 00:03:29 +0900
> From: shanice msuya <shanicemsuya at gmail.com>
> Subject: Re: [Disksim-users] Help with Segmentation Fault
> To: disksim-users at ece.cmu.edu
> Message-ID: <BANLkTimNa-g8n2vS30MzJ1V0_BbvXh_XOw at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hello,
> Please somebody help me with known tools which can generate block I/O
> synthetic workloads for open and closed models.
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://sos.ece.cmu.edu/pipermail/disksim-users/attachments/20110508/0b4e49be/attachment-0001.html
> >
>
> ------------------------------
>
> Message: 2
> Date: Sun, 8 May 2011 00:04:47 +0900
> From: shanice msuya <shanicemsuya at gmail.com>
> Subject: [Disksim-users] block I/O synthetic workload tools
> To: disksim-users at ece.cmu.edu
> Message-ID: <BANLkTinDga-GUirjNaEOTKqDmYRTmSxTmQ at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hello,
> Please somebody help me with known tools which can generate block I/O
> synthetic workloads for open and closed models.
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://sos.ece.cmu.edu/pipermail/disksim-users/attachments/20110508/e7c92a44/attachment-0001.html
> >
>
> ------------------------------
>
> Message: 3
> Date: Sat, 7 May 2011 19:39:29 +0430
> From: Alireza Haghdoost <haghdoost at gmail.com>
> Subject: Re: [Disksim-users] block I/O synthetic workload tools
> To: shanice msuya <shanicemsuya at gmail.com>
> Cc: disksim-users at ece.cmu.edu
> Message-ID: <BANLkTik-S2h7ee_x9Vwg7WP8cDe97E7Pig at mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
>
> postmark, iozone, intel open storage toolkit , vdbench ...
>
>
> Alireza Haghdoost
> Graduate Student
> Department of Computer Engineering
> Sharif University of Technology
> http://ce.sharif.edu/~haghdoost
> Phone: (98) (21) 6616 6689
>
>
>
>
> 2011/5/7 shanice msuya <shanicemsuya at gmail.com>
>
> > Hello,
> > Please somebody help me with known tools which can generate block I/O
> > synthetic workloads for open and closed models.
> > _______________________________________________
> > Disksim-users mailing list
> > Disksim-users at ece.cmu.edu
> > https://sos.ece.cmu.edu/mailman/listinfo/disksim-users
> >
> >
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://sos.ece.cmu.edu/pipermail/disksim-users/attachments/20110507/4e9a8a4e/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 67, Issue 3
> ********************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110508/7ce397cc/attachment.html>

From chensinhome at gmail.com  Sat May  7 18:34:36 2011
From: chensinhome at gmail.com (=?Big5?B?q0inu7Ov?=)
Date: Sun, 8 May 2011 06:34:36 +0800
Subject: [Disksim-users] Disksim-users Digest, Vol 66, Issue 6
In-Reply-To: <mailman.1.1302364804.20757.disksim-users@ece.cmu.edu>
References: <mailman.1.1302364804.20757.disksim-users@ece.cmu.edu>
Message-ID: <BANLkTim=4YkGiQD4iLJmvhTiPg2s0dqc0Q@mail.gmail.com>

2011/4/10 <disksim-users-request at ece.cmu.edu>

> 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. parallel flash chips (f.rajaei)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Sat, 9 Apr 2011 15:57:37 +0430
> From: "f.rajaei" <frajaei55 at gmail.com>
> Subject: [Disksim-users] parallel flash chips
> To: disksim-users at ece.cmu.edu
> Message-ID: <BANLkTikyZ6NmtxOba2wQzW3VsbUobRAhgQ at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Dear All,
>
> Does anybody know if it is possible to write or read on an SSD emelents
> (flash chips) in parallel fasion (striping) since simulation results show
> different number of writes or reads on different elements of an SSD.
>


   yes ,  it is possible.

   if a flash chip consist of multiple die which can perform IO
independently


>
> Regards,
> Farzaneh
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://sos.ece.cmu.edu/pipermail/disksim-users/attachments/20110409/354271c4/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 66, Issue 6
> ********************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110508/4e827bcf/attachment.html>

From shanicemsuya at gmail.com  Mon May  9 11:22:31 2011
From: shanicemsuya at gmail.com (shanice msuya)
Date: Tue, 10 May 2011 00:22:31 +0900
Subject: [Disksim-users] RAID 6 Configuration in DiskSim
Message-ID: <BANLkTi=f5ufD8DsHJYprdAREypRGLv5vHA@mail.gmail.com>

Hello disksim users,
I  read the manual, I understand how to configure RAID 5,
e.g

Addressing mode = Array,

Distribution scheme = Striped,

Redundancy scheme = Parity_rotated,

devices = [ disk0 .. disk8 ]


but I am not sure how to configure RAID 6. I understand RAID 6 need two
independent parity schemes.
Can somebody help me
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110510/34c69570/attachment.html>

From haghdoost at gmail.com  Mon May  9 15:49:18 2011
From: haghdoost at gmail.com (Alireza Haghdoost)
Date: Tue, 10 May 2011 00:19:18 +0430
Subject: [Disksim-users] RAID 6 Configuration in DiskSim
In-Reply-To: <BANLkTi=f5ufD8DsHJYprdAREypRGLv5vHA@mail.gmail.com>
References: <BANLkTi=f5ufD8DsHJYprdAREypRGLv5vHA@mail.gmail.com>
Message-ID: <BANLkTikf_vgkayWvgtv35p1XKgVG9mnXcQ@mail.gmail.com>

There is no RAID 6 configuration in Disksim.

Alireza Haghdoost
Graduate Student
Department of Computer Engineering
Sharif University of Technology
http://ce.sharif.edu/~haghdoost
Phone: (98) (21) 6616 6689





2011/5/9 shanice msuya <shanicemsuya at gmail.com>:
> Hello disksim users,
> I? read the manual, I understand how to configure RAID 5,
> e.g
>
> Addressing mode = Array,
>
> Distribution scheme = Striped,
>
> Redundancy scheme = Parity_rotated,
>
> devices = [ disk0 .. disk8 ]
>
> but I am not sure how to configure RAID 6. I understand RAID 6 need two
> independent parity schemes.
> Can somebody help me
>
> _______________________________________________
> Disksim-users mailing list
> Disksim-users at ece.cmu.edu
> https://sos.ece.cmu.edu/mailman/listinfo/disksim-users
>
>


From 258908913 at qq.com  Fri May 13 09:07:35 2011
From: 258908913 at qq.com (=?gbk?B?c21pbGU=?=)
Date: Fri, 13 May 2011 21:07:35 +0800
Subject: [Disksim-users] RAID 5 Configurations for SSD in Disksim --help
Message-ID: <tencent_1343FAF756328F9C176E1450@qq.com>

hi,disksim users
     By reading the manual, I know disksim support RAID 5 for SSD.But either  in the ssd_iozone.parv or the ssd-sw5m.parv, there is only one ssd. We  know that if we want to  configure RAID 5,we need three or more SSDs.

     should I modify the topology? I tried,but failed. Can you help me? 
    I modified the ssd-sw5m.parv as follows:(only  list the modified part)
    ssdmodel_ssd SSD0 {
     ??
     }
    ssdmodel_ssd SSD1 {
      ??
      }
    ssdmodel_ssd SSD2 {
      ??
      }
    ssdmodel_ssd SSD3 {
      ??
      }
    instantiate [ ssd0x0 ] as  SSD0
    instantiate [ ssd0x1 ] as  SSD1
    instantiate [ ssd0x2 ] as  SSD2
    instantiate [ ssd0x3 ] as  SSD3

# system topology
topology disksim_iodriver driver0 [
     disksim_bus bustop [

          ############## HBA 0 #############################
          disksim_ctlr ctlrHBA0 [
               disksim_bus busHBA0 [
                    ############## Flash-device Raid Controller ###############
                    ssdmodel_ssd ssd0x0 [],
                    ssdmodel_ssd ssd0x1 [],
                    ssdmodel_ssd ssd0x2 [],
                    ssdmodel_ssd ssd0x3 []    
                 ]     # end of bus0
            ]        # end of HBA0

          ############ INSERT MORE HBA's######################          
            
     ]           # end of bustop
]              # end of driver0 (and system topology)

# four disksim_logorg  from org0 to org3
disksim_logorg org0 {
   Addressing mode = Array,
   Distribution scheme = Striped,
   Redundancy scheme = Parity_rotated,

   # vp - added more ssd elements
   devices = [ ssd0x0 ],
      ??
} # end of logorg org0 spec

disksim_logorg org1 {
    Addressing mode = Array,
    Distribution scheme = Striped,
    Redundancy scheme = Parity_rotated,
 
    # vp - added more ssd elements
    devices = [ ssd0x1 ],
       ??
 } # end of logorg org1 spec

disksim_logorg org2 {
    Addressing mode = Array,
    Distribution scheme = Striped,
    Redundancy scheme = Parity_rotated,
 
    # vp - added more ssd elements
    devices = [ ssd0x2 ],
       ??
 } # end of logorg org2 spec

disksim_logorg org3 {
    Addressing mode = Array,
    Distribution scheme = Striped,
    Redundancy scheme = Parity_rotated,
 
    # vp - added more ssd elements
    devices = [ ssd0x3 ],
       ??
 } # end of logorg org3 spec

#four generators from 0 to 3
disksim_synthgen { # generator 0
   Storage capacity per device  =  7018904,
   devices = [ org0 ],
   ??
},# end of generator 1
disksim_synthgen { # generator 1
   Storage capacity per device  =  7018904,
   devices = [ org1 ],
   ??
}, # end of generator 2
disksim_synthgen { # generator 2
   Storage capacity per device  =  7018904,
   devices = [ org2 ],
   ??
}, # end of generator 2
disksim_synthgen { # generator 3
   Storage capacity per device  =  7018904,
   devices = [ org3 ],
   ??
} # end of generator 3

    above is the content I changed ,but these errors appeared.

unexpected request location: devno 3, blkno 779696, bcount 8

Assertion failed:
simtime     = 0.744525
totalreqs     = 1
disksim: disksim_logorg.c:763: logorg_maprequest: Assertion `logorgno != -1' failed.
???
     I don't know why,so please help!
     Best regards!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110513/99ae5dec/attachment.html>

From liuzhuo9090 at gmail.com  Fri May 13 12:48:12 2011
From: liuzhuo9090 at gmail.com (liu zhuo)
Date: Fri, 13 May 2011 10:48:12 -0600
Subject: [Disksim-users] devno, blkno in ascii trace and diskno in log_org
Message-ID: <BANLkTimNgVU7RwdbGWmYEUcdj3OWbQGDgA@mail.gmail.com>

Hello everyone.

I want to know about the relationship between the devno in ascii trace
(second colume) and the diskno in logical organization.
For example, assuming that there's only one log_org configured in the
parameter file, with several disks organized in "Parts".
How does the devno in the ascii trace entry map to the diskno in the
log_org?
Is the blkno dedicated for the device or disk with the devno?  I mean, do
disks with different devno have seperate block address space (blkno)?
Thanks, sincerely

-- 
Zhuo Liu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110513/b3ff2572/attachment.html>

From shanicemsuya at gmail.com  Sun May 15 09:42:46 2011
From: shanicemsuya at gmail.com (shanice msuya)
Date: Sun, 15 May 2011 22:42:46 +0900
Subject: [Disksim-users] No 'tracedate' line in HPL trace
Message-ID: <BANLkTimtZSXSwbEpLjQDd4FjbKVJdb2Jog@mail.gmail.com>

Hi,
I keep getting the error "No 'tracedate' line in HPL trace" when I use
cello99 traces.
Anyone with an idea on how to resolve the issue?



Shanice.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110515/2002b122/attachment.html>

From 258908913 at qq.com  Tue May 17 05:53:30 2011
From: 258908913 at qq.com (=?ISO-8859-1?B?c21pbGU=?=)
Date: Tue, 17 May 2011 17:53:30 +0800
Subject: [Disksim-users] How to Simulate RAID5 based on one ssd?
Message-ID: <tencent_76E3FCA02D807FE347A4DCD9@qq.com>

hi,disksim users
       If I want to simulate RAID5 based on one ssd,maybe I should modify the ssd.c  and the related codes,it is so terrible. Can I transfer the question to this case:
      A big ssd with 8 elements and its total capacity is 8G, if I want to simulate RAID5 based on this ssd.Simulating RAID5 based on one ssd is not supported in Disksim,but n ssd can do in Disksim.
       So,can I transfer the question to this :
       Simulating RAID5 based on 8 small ssds,and every ssd with only one element,the total capacity of 8 ssds is 8G,too.We should ensure the total capacity is the same.then I can use Disksim to simulate RAID5.
       In first case,we use parallel between elements.and the second case ,we use parallel between ssds(here ssds equal elements,because one ssd with only one element).
       Are the two parallels the same?   Are they equivalent?   Is it feasible?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110517/43ee4dec/attachment.html>

From member at linkedin.com  Wed May 18 18:36:55 2011
From: member at linkedin.com (Zhuo Liu via LinkedIn)
Date: Wed, 18 May 2011 22:36:55 +0000 (UTC)
Subject: [Disksim-users] Invitation to connect on LinkedIn
Message-ID: <398935896.20549579.1305758215205.JavaMail.app@ela4-bed83.prod>

LinkedIn
------------Zhuo Liu requested to add you as a connection on LinkedIn:
------------------------------------------

Kai,

I'd like to add you to my professional network on LinkedIn.

- Zhuo

Accept invitation from Zhuo Liu
http://www.linkedin.com/e/-ljq2hr-gnuuq71f-2p/T3ggIQCp7GGsDtIGS9QgX8CC7RpBvymv1SWjuxY/blk/I1346846826_3/1BpC5vrmRLoRZcjkkZt5YCpnlOt3RApnhMpmdzgmhxrSNBszYPnPoOe3oQe3oQcP59bRwRlmFCrRZdbPoOdjsUdzsOej8LrCBxbOYWrSlI/EML_comm_afe/

View invitation from Zhuo Liu
http://www.linkedin.com/e/-ljq2hr-gnuuq71f-2p/T3ggIQCp7GGsDtIGS9QgX8CC7RpBvymv1SWjuxY/blk/I1346846826_3/3dvdz8UdzgUdzgPckALqnpPbOYWrSlI/svi/

------------------------------------------ 
DID YOU KNOW your LinkedIn profile helps you control your public image when people search for you? Setting your profile as public means your LinkedIn profile will come up when people enter your name in leading search engines. Take control of your image! 
http://www.linkedin.com/e/-ljq2hr-gnuuq71f-2p/ewp/inv-22/

 
-- 
(c) 2011, LinkedIn Corporation
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110518/2145a24c/attachment.html>

From adamcrume at gmail.com  Thu May 19 18:53:24 2011
From: adamcrume at gmail.com (Adam Crume)
Date: Thu, 19 May 2011 15:53:24 -0700
Subject: [Disksim-users] RAID 5
Message-ID: <BANLkTinrjeTk3+my0qFzM2D-uiC9Sh7Agg@mail.gmail.com>

How well tested is RAID 5 in DiskSim?  I set up a test to look at the
average request time for writes, varying the write size and offset
into a stripe.  I'm getting strange results.  For unaligned writes,
the average time drops when the write size reaches one stripe, even
though it is really two partial stripes.  For aligned writes, the
average time doesn't drop until one stripe width plus one sector.


From ffalanga at fastwebnet.it  Sat May 21 04:37:33 2011
From: ffalanga at fastwebnet.it (Francesco Falanga)
Date: Sat, 21 May 2011 10:37:33 +0200
Subject: [Disksim-users] Closed storage subsytem models
Message-ID: <9B5BCBFA06224F32AE5AD3FF48350CF5@numonyx.local>

Dear disksim Users,

I am trying for so long to configure disksim so that external workload traces are handled in a closed way ( :a new request is picked up from the trace list and launched only when previous command execution is finished).
So far I was only able to configure this way the synth worload generator but I am not able to apply this configuration to external traces.

I read in some documentation that it should be possible to configure also the 'closed' behavior for external trace but cannot find how.

Can someone help me?

Thank you a lot & Regards,

Francesco. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110521/3eb2b0ff/attachment.html>

From chensinhome at gmail.com  Sat May 21 18:03:17 2011
From: chensinhome at gmail.com (=?Big5?B?q0inu7Ov?=)
Date: Sun, 22 May 2011 06:03:17 +0800
Subject: [Disksim-users] Disksim-users Digest, Vol 67, Issue 13
In-Reply-To: <mailman.1.1305993604.17558.disksim-users@ece.cmu.edu>
References: <mailman.1.1305993604.17558.disksim-users@ece.cmu.edu>
Message-ID: <BANLkTi=RF-SgCDV_hPdTx_zFFsKsTcwOew@mail.gmail.com>

what is your error massages?


2011/5/22 <disksim-users-request at ece.cmu.edu>

> 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. Closed storage subsytem models (Francesco Falanga)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Sat, 21 May 2011 10:37:33 +0200
> From: "Francesco Falanga" <ffalanga at fastwebnet.it>
> Subject: [Disksim-users] Closed storage subsytem models
> To: <disksim-users at ece.cmu.edu>
> Message-ID: <9B5BCBFA06224F32AE5AD3FF48350CF5 at numonyx.local>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Dear disksim Users,
>
> I am trying for so long to configure disksim so that external workload
> traces are handled in a closed way ( :a new request is picked up from the
> trace list and launched only when previous command execution is finished).
> So far I was only able to configure this way the synth worload generator
> but I am not able to apply this configuration to external traces.
>
> I read in some documentation that it should be possible to configure also
> the 'closed' behavior for external trace but cannot find how.
>
> Can someone help me?
>
> Thank you a lot & Regards,
>
> Francesco.
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://sos.ece.cmu.edu/pipermail/disksim-users/attachments/20110521/3eb2b0ff/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 67, Issue 13
> *********************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110522/2c244ede/attachment.html>

From poohtimu at gmail.com  Mon May 23 12:50:16 2011
From: poohtimu at gmail.com (poohtimu)
Date: Tue, 24 May 2011 00:50:16 +0800
Subject: [Disksim-users] ssd-add-on questions
Message-ID: <1306169416.27548.0.camel@poohtimu-linux>

Hi all,
I have some question about using ssd-add-on.
I have made it work properly in default setting while
runvalid(/ssdmodel/runvalid)
but while I modified the parameter in ssd spec, "Plane block mapping"
and "Number of parallel units", there is something wrong.
when I set Plane block mapping = 1 and Number of parallel units = 2,
synthetic workload "rw5m" and real traces "iozone" will generate
segmentation fault.
moreover, Plane block mapping = 3 and Number of parallel units = 2 
          Plane block mapping = 1 and Number of parallel units = 4 
          Plane block mapping = 3 and Number of parallel units = 4 
these parameter has same result. 
    
All I did is modified this two parameter, I can't figure out why there
is a segmentation fault
Is there anything I missed?
Please help me, I will appreciate your help

Best Regards



From ericoolbe at ymail.com  Tue May 24 07:27:34 2011
From: ericoolbe at ymail.com (YaoQing,Wang)
Date: Tue, 24 May 2011 19:27:34 +0800
Subject: [Disksim-users] ssd-add-on questions
In-Reply-To: <1306169416.27548.0.camel@poohtimu-linux>
References: <1306169416.27548.0.camel@poohtimu-linux>
Message-ID: <BANLkTi=2h=UqaTgmdW-ocBSLqDjLR1_haQ@mail.gmail.com>

hi~

maybe try to add a line in the function of " ssd_media_access_request_element
"
original:
tmp->blkno = tmp->blkno;

modified:
tmp->blkno = tmp->blkno;
tmp->opid = curr->opid;

good luck

On Tue, May 24, 2011 at 12:50 AM, poohtimu <poohtimu at gmail.com> wrote:

> Hi all,
> I have some question about using ssd-add-on.
> I have made it work properly in default setting while
> runvalid(/ssdmodel/runvalid)
> but while I modified the parameter in ssd spec, "Plane block mapping"
> and "Number of parallel units", there is something wrong.
> when I set Plane block mapping = 1 and Number of parallel units = 2,
> synthetic workload "rw5m" and real traces "iozone" will generate
> segmentation fault.
> moreover, Plane block mapping = 3 and Number of parallel units = 2
>          Plane block mapping = 1 and Number of parallel units = 4
>          Plane block mapping = 3 and Number of parallel units = 4
> these parameter has same result.
>
> All I did is modified this two parameter, I can't figure out why there
> is a segmentation fault
> Is there anything I missed?
> Please help me, I will appreciate your help
>
> Best Regards
>
> _______________________________________________
> Disksim-users mailing list
> Disksim-users at ece.cmu.edu
> https://sos.ece.cmu.edu/mailman/listinfo/disksim-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110524/24df04a4/attachment.html>

From poohtimu at gmail.com  Wed May 25 03:37:06 2011
From: poohtimu at gmail.com (=?Big5?B?uK2sZqao?=)
Date: Wed, 25 May 2011 15:37:06 +0800
Subject: [Disksim-users] Disksim-users Digest, Vol 67, Issue 15
In-Reply-To: <mailman.1.1306252809.23877.disksim-users@ece.cmu.edu>
References: <mailman.1.1306252809.23877.disksim-users@ece.cmu.edu>
Message-ID: <BANLkTim-1fFuw1Vx=1=PdA+7eRt8NaSPKA@mail.gmail.com>

Hi,

It really works!! Including Number of parallel units = 1,2,4
But original should be "tmp->blkno = blkno"; not "tmp->blkno = tmp->blkno;"
 Thanks anyway. :D
Now I have a new problem about saturation msg while Plane block mapping =
simple concatenation in postmark trace
I will try first, thank you very much.

2011/5/25 <disksim-users-request at ece.cmu.edu>

> 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. ssd-add-on questions (poohtimu)
>   2. Re: ssd-add-on questions (YaoQing,Wang)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Tue, 24 May 2011 00:50:16 +0800
> From: poohtimu <poohtimu at gmail.com>
> Subject: [Disksim-users] ssd-add-on questions
> To: disksim-users at ece.cmu.edu
> Message-ID: <1306169416.27548.0.camel at poohtimu-linux>
> Content-Type: text/plain; charset="UTF-8"
>
> Hi all,
> I have some question about using ssd-add-on.
> I have made it work properly in default setting while
> runvalid(/ssdmodel/runvalid)
> but while I modified the parameter in ssd spec, "Plane block mapping"
> and "Number of parallel units", there is something wrong.
> when I set Plane block mapping = 1 and Number of parallel units = 2,
> synthetic workload "rw5m" and real traces "iozone" will generate
> segmentation fault.
> moreover, Plane block mapping = 3 and Number of parallel units = 2
>          Plane block mapping = 1 and Number of parallel units = 4
>          Plane block mapping = 3 and Number of parallel units = 4
> these parameter has same result.
>
> All I did is modified this two parameter, I can't figure out why there
> is a segmentation fault
> Is there anything I missed?
> Please help me, I will appreciate your help
>
> Best Regards
>
>
>
> ------------------------------
>
> Message: 2
> Date: Tue, 24 May 2011 19:27:34 +0800
> From: "YaoQing,Wang" <ericoolbe at ymail.com>
> Subject: Re: [Disksim-users] ssd-add-on questions
> To: disksim-users at ece.cmu.edu
> Message-ID: <BANLkTi=2h=UqaTgmdW-ocBSLqDjLR1_haQ at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> hi~
>
> maybe try to add a line in the function of "
> ssd_media_access_request_element
> "
> original:
> tmp->blkno = tmp->blkno;
>
> modified:
> tmp->blkno = tmp->blkno;
> tmp->opid = curr->opid;
>
> good luck
>
> On Tue, May 24, 2011 at 12:50 AM, poohtimu <poohtimu at gmail.com> wrote:
>
> > Hi all,
> > I have some question about using ssd-add-on.
> > I have made it work properly in default setting while
> > runvalid(/ssdmodel/runvalid)
> > but while I modified the parameter in ssd spec, "Plane block mapping"
> > and "Number of parallel units", there is something wrong.
> > when I set Plane block mapping = 1 and Number of parallel units = 2,
> > synthetic workload "rw5m" and real traces "iozone" will generate
> > segmentation fault.
> > moreover, Plane block mapping = 3 and Number of parallel units = 2
> >          Plane block mapping = 1 and Number of parallel units = 4
> >          Plane block mapping = 3 and Number of parallel units = 4
> > these parameter has same result.
> >
> > All I did is modified this two parameter, I can't figure out why there
> > is a segmentation fault
> > Is there anything I missed?
> > Please help me, I will appreciate your help
> >
> > Best Regards
> >
> > _______________________________________________
> > Disksim-users mailing list
> > Disksim-users at ece.cmu.edu
> > https://sos.ece.cmu.edu/mailman/listinfo/disksim-users
> >
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://sos.ece.cmu.edu/pipermail/disksim-users/attachments/20110524/24df04a4/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 67, Issue 15
> *********************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110525/5d757b07/attachment.html>

From ffalanga at fastwebnet.it  Thu May 26 11:42:30 2011
From: ffalanga at fastwebnet.it (ffalanga at fastwebnet.it)
Date: Thu, 26 May 2011 17:42:30 +0200 (CEST)
Subject: [Disksim-users] R: Re:  Disksim-users Digest, Vol 67, Issue 13
Message-ID: <14002919.1333721306424550698.JavaMail.root@pr003msr>

Error is 'saturation'.
The reason is that I remove command queuing (queue len==1) so the time when next
 command arrive the previous one is still there while being finished.
The point is I would the simulator not issue the command basing upon the arrival
 time of each command (indicated into the external trace) but rather simply I wa
nt the simulator to issue next command in the list only when the previous is end
ed.

I need to evaluate the performances of the device in executing a specific comman
ds list with no queue in the system.

Francesco.   

<br>
<br>
<blockquote>
----Messaggio originale----<br>
Da: chensinhome at gmail.com<br>
Data: 22/05/2011 0.03<br>
A: &lt;disksim-users at ece.cmu.edu&gt;<br>
Ogg: Re: [Disksim-users] Disksim-users Digest, Vol 67, Issue 13<br>
<br>
<div>what is your error massages?</div><br><br><div class="gmail_quote">2011/5/2
2  <span dir="ltr">&lt;<a href="mailto:disksim-users-request at ece.cmu.edu">disksi
m-users-request at ece.cmu.edu</a>&gt;</span><br><blockquote class="gmail_quote" st
yle="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
Send Disksim-users mailing list submissions to<br>
 ? ? ? ?<a href="mailto:disksim-users at ece.cmu.edu">disksim-users at ece.cmu.edu</a>
<br>
<br>
To subscribe or unsubscribe via the World Wide Web, visit<br>
 ? ? ? ?<a href="https://sos.ece.cmu.edu/mailman/listinfo/disksim-users" target=
"_blank">https://sos.ece.cmu.edu/mailman/listinfo/disksim-users</a><br>
or, via email, send a message with subject or body &#39;help&#39; to<br>
 ? ? ? ?<a href="mailto:disksim-users-request at ece.cmu.edu">disksim-users-request
@ece.cmu.edu</a><br>
<br>
You can reach the person managing the list at<br>
 ? ? ? ?<a href="mailto:disksim-users-owner at ece.cmu.edu">disksim-users-owner at ece
.cmu.edu</a><br>
<br>
When replying, please edit your Subject line so it is more specific<br>
than &quot;Re: Contents of Disksim-users digest...&quot;<br>
<br>
<br>
Today&#39;s Topics:<br>
<br>
 ? 1. Closed storage subsytem models (Francesco Falanga)<br>
<br>
<br>
----------------------------------------------------------------------<br>
<br>
Message: 1<br>
Date: Sat, 21 May 2011 10:37:33 +0200<br>
From: &quot;Francesco Falanga&quot; &lt;<a href="mailto:ffalanga at fastwebnet.it">
ffalanga at fastwebnet.it</a>&gt;<br>
Subject: [Disksim-users] Closed storage subsytem models<br>
To: &lt;<a href="mailto:disksim-users at ece.cmu.edu">disksim-users at ece.cmu.edu</a>
&gt;<br>
Message-ID: &lt;9B5BCBFA06224F32AE5AD3FF48350CF5 at numonyx.local&gt;<br>
Content-Type: text/plain; charset=&quot;iso-8859-1&quot;<br>
<br>
Dear disksim Users,<br>
<br>
I am trying for so long to configure disksim so that external workload traces ar
e handled in a closed way ( :a new request is picked up from the trace list and 
launched only when previous command execution is finished).<br>

So far I was only able to configure this way the synth worload generator but I a
m not able to apply this configuration to external traces.<br>
<br>
I read in some documentation that it should be possible to configure also the &#
39;closed&#39; behavior for external trace but cannot find how.<br>
<br>
Can someone help me?<br>
<br>
Thank you a lot &amp; Regards,<br>
<br>
Francesco.<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: &lt;<a href="http://sos.ece.cmu.edu/pipermail/disksim-users/attachments/201
10521/3eb2b0ff/attachment-0001.html" target="_blank">http://sos.ece.cmu.edu/pipe
rmail/disksim-users/attachments/20110521/3eb2b0ff/attachment-0001.html</a>&gt;<b
r>

<br>
------------------------------<br>
<br>
_______________________________________________<br>
Disksim-users mailing list<br>
<a href="mailto:Disksim-users at ece.cmu.edu">Disksim-users at ece.cmu.edu</a><br>
<a href="https://sos.ece.cmu.edu/mailman/listinfo/disksim-users" target="_blank"
>https://sos.ece.cmu.edu/mailman/listinfo/disksim-users</a><br>
<br>
<br>
End of Disksim-users Digest, Vol 67, Issue 13<br>
*********************************************<br>
</blockquote></div><br>
<br>
</blockquote><br>



From 258908913 at qq.com  Fri May 27 05:15:12 2011
From: 258908913 at qq.com (=?ISO-8859-1?B?c21pbGU=?=)
Date: Fri, 27 May 2011 17:15:12 +0800
Subject: [Disksim-users] how to find the entrance of the read or write
	request served based on one ssd
Message-ID: <tencent_2BC5CDCA075DDFC01CFC71D9@qq.com>

Hi,
     I print the function it executes,but so many function make me confused. I just want to know the entrance of the read or write request processing based on one ssd,then I can cut it into several parts to issue on differerent elements. My goal is to improve the bandwidth.
     Can you help me?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20110527/b34ab7bf/attachment.html>