From mohan at cs.virginia.edu  Tue Jan  6 14:15:59 2009
From: mohan at cs.virginia.edu (Bhushan)
Date: Tue, 6 Jan 2009 14:15:59 -0500
Subject: [Disksim-users] Response Time for each request.
Message-ID: <3330de3a0901061115i529cfb29xd1bf4731a3f875b1@mail.gmail.com>

Hi,
I am using the SSD Add-on for Disksim analyzing the performance of different
traces. Is there a way to determine the response time for each
request(read/write) in the trace? While debugging through the code, I found
the "ioqueue_remove_completed_request" function call in disksim-ioqueue.c,
while updates the "outtimestats" variable with the response time but the
problem is this function gets called multiple times for each request.
I need this functionality to understand how the SSD responds for different
requests.

Any help would be greatly appreciated.
-- 
Regards,
Bhushan.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20090106/5d29c9ef/attachment.html>

From wobber at microsoft.com  Tue Jan  6 16:04:37 2009
From: wobber at microsoft.com (Ted Wobber)
Date: Tue, 6 Jan 2009 13:04:37 -0800
Subject: [Disksim-users] Response Time for each request.
In-Reply-To: <3330de3a0901061115i529cfb29xd1bf4731a3f875b1@mail.gmail.com>
References: <3330de3a0901061115i529cfb29xd1bf4731a3f875b1@mail.gmail.com>
Message-ID: <338B06DAAFBCFD4894EFCAC38E7B811A2035C6BA5E@NA-EXMSG-C113.redmond.corp.microsoft.com>

Bhushan -

Assuming that you are happy with a statistical summary of requests, the SSD add-on computes the request latency in aggregate and also gives information about the distribution of reads and writes.

It also gives similar numbers for each simulated flash chip.

If you need more detailed output, you can always hack the code.  The SSD request completion handler is in ssd.c:  ssd_request_complete.  There are also chip (element) access completion routines in that module.

Best regards,

Ted Wobber
MSR Silicon Valley


From: disksim-users-bounces at ece.cmu.edu [mailto:disksim-users-bounces at ece.cmu.edu] On Behalf Of Bhushan
Sent: Tuesday, January 06, 2009 11:16 AM
To: disksim-users
Subject: [Disksim-users] Response Time for each request.

Hi,
I am using the SSD Add-on for Disksim analyzing the performance of different traces. Is there a way to determine the response time for each request(read/write) in the trace? While debugging through the code, I found the "ioqueue_remove_completed_request" function call in disksim-ioqueue.c, while updates the "outtimestats" variable with the response time but the problem is this function gets called multiple times for each request.
I need this functionality to understand how the SSD responds for different requests.

Any help would be greatly appreciated.
--
Regards,
Bhushan.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20090106/43913275/attachment.html>

From mohan at cs.virginia.edu  Wed Jan  7 18:07:34 2009
From: mohan at cs.virginia.edu (Bhushan)
Date: Wed, 7 Jan 2009 18:07:34 -0500
Subject: [Disksim-users] Response Time for each request.
In-Reply-To: <338B06DAAFBCFD4894EFCAC38E7B811A2035C6BA5E@NA-EXMSG-C113.redmond.corp.microsoft.com>
References: <3330de3a0901061115i529cfb29xd1bf4731a3f875b1@mail.gmail.com>
	<338B06DAAFBCFD4894EFCAC38E7B811A2035C6BA5E@NA-EXMSG-C113.redmond.corp.microsoft.com>
Message-ID: <3330de3a0901071507t2d3198c2ga0466676f952a51f@mail.gmail.com>

Thanks Ted. I was looking at disksim_ioqueue.c
"ioqueue_remove_completed_request" instead of ssd.c. Your hint helped me
find what I needed. Thanks again.

On Tue, Jan 6, 2009 at 4:04 PM, Ted Wobber <wobber at microsoft.com> wrote:

>  Bhushan ?
>
>
>
> Assuming that you are happy with a statistical summary of requests, the SSD
> add-on computes the request latency in aggregate and also gives information
> about the distribution of reads and writes.
>
>
>
> It also gives similar numbers for each simulated flash chip.
>
>
>
> If you need more detailed output, you can always hack the code.  The SSD
> request completion handler is in ssd.c:  ssd_request_complete.  There are
> also chip (element) access completion routines in that module.
>
>
>
> Best regards,
>
>
>
> Ted Wobber
>
> MSR Silicon Valley
>
>
>
>
>
> *From:* disksim-users-bounces at ece.cmu.edu [mailto:
> disksim-users-bounces at ece.cmu.edu] *On Behalf Of *Bhushan
> *Sent:* Tuesday, January 06, 2009 11:16 AM
> *To:* disksim-users
> *Subject:* [Disksim-users] Response Time for each request.
>
>
>
> Hi,
> I am using the SSD Add-on for Disksim analyzing the performance of
> different traces. Is there a way to determine the response time for each
> request(read/write) in the trace? While debugging through the code, I found
> the "ioqueue_remove_completed_request" function call in disksim-ioqueue.c,
> while updates the "outtimestats" variable with the response time but the
> problem is this function gets called multiple times for each request.
> I need this functionality to understand how the SSD responds for different
> requests.
>
> Any help would be greatly appreciated.
> --
> Regards,
> Bhushan.
>



-- 
Regards,
Bhushan.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20090107/63946700/attachment.html>

From admin at rodale.com  Mon Jan 12 03:42:21 2009
From: admin at rodale.com (admin at rodale.com)
Date: Mon, 12 Jan 2009 03:42:21 -0500 (EST)
Subject: [Disksim-users] Men's Health News(Medical College of Tennessee )
Message-ID: <3902346237.3173599775270.JavaMail.wasuser@was-colo-05>

An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20090112/56ced6da/attachment.html>

From lisaliushu at gmail.com  Sun Jan 18 22:17:55 2009
From: lisaliushu at gmail.com (lisa)
Date: Mon, 19 Jan 2009 11:17:55 +0800
Subject: [Disksim-users] disksim with power model
Message-ID: <3ab709a40901181917i468e48bld31b169636dc8070@mail.gmail.com>

Hi all,

Does anybody know of a source code release for DiskSim that adds power
models? There was some discussion about Power Model in the maillist in 2007.
However, no more information can be found. If anyone has, I was wondering if
you can send the code.  Thanks very much.

Lisa
2009.1.19
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20090119/f70e1652/attachment.html>

From rimuni at gmail.com  Mon Jan 19 16:36:34 2009
From: rimuni at gmail.com (Maliha)
Date: Mon, 19 Jan 2009 13:36:34 -0800
Subject: [Disksim-users] Compile error - duplicate case value
Message-ID: <949d418b0901191336u4e4fd9e6l899bef6c5a003f94@mail.gmail.com>

Hi

I got the following error while trying to compile Disksim on my machine:

disksim_iosim.c: In function 'io_initialize':
disksim_iosim.c:712: error: duplicate case value
disksim_iosim.c:712: error: previously used here
make[1]: *** [disksim_iosim.o] Error 1

I was able to compile the code by increasing DISKSIM_EVENT_SIZE, but
was not able to run it; Disksim showed "Segmentation Fault".

Please let me know what to do.

Thanks,
Maliha


From ganger at ece.cmu.edu  Mon Jan 19 19:32:43 2009
From: ganger at ece.cmu.edu (Greg Ganger)
Date: Mon, 19 Jan 2009 19:32:43 -0500 (EST)
Subject: [Disksim-users] disksim with power model
In-Reply-To: <3ab709a40901181917i468e48bld31b169636dc8070@mail.gmail.com>
References: <3ab709a40901181917i468e48bld31b169636dc8070@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0901191932000.19876@nirvana.pdl.cmu.edu>


I don't, although I think that there was a project at Princeton (under
Randy Wang) that built/explored such modeling.  I don't know if they
released their code.

Greg


On Mon, 19 Jan 2009, lisa wrote:

> Hi all,
>
> Does anybody know of a source code release for DiskSim that adds power
> models? There was some discussion about Power Model in the maillist in 2007.
> However, no more information can be found. If anyone has, I was wondering if
> you can send the code.  Thanks very much.
>
> Lisa
> 2009.1.19
>


From rimuni at gmail.com  Fri Jan 23 17:36:10 2009
From: rimuni at gmail.com (Maliha)
Date: Fri, 23 Jan 2009 14:36:10 -0800
Subject: [Disksim-users]  segmentation fault while simulating ssd
Message-ID: <949d418b0901231436j49825a93tef7dda3e2fa01059@mail.gmail.com>

Hi

I am having segmentation faults while using disksim for simulating
ssd. Disksim works fine if it is compiled without the ssd feature, but
with the ssd add-on it gives segmentation fault for each of the
simulations in 'runvalid'.

please let me know If anyone has the solution.

thanks,
maliha