From 6375625 at gmail.com  Wed Jun 12 00:03:44 2013
From: 6375625 at gmail.com (q q)
Date: Tue, 11 Jun 2013 22:03:44 -0600
Subject: [Disksim-users] A question about time-limited request
Message-ID: <CAALB-o8MrZt9G8EBpN7e4req0OLVhpqFvrGHEbLbqeNJnwt2Gw@mail.gmail.com>

If a time-limited request is serviced before its time limit, does the
generator process block and wait till the end of the time limit before
generating the next request?

Thank you.

On Page 42 of the Disksim 4.0 manual,

Probability of time-limited request
"This specifies the probability that a generated request is
time-limited. That is, the corresponding generator process ?blocks?
and waits for the request to complete (if it is not already complete)
after a given amount of think time (specified by the below ?time
limit? parameters) [5, 2]."


From ganger at ece.cmu.edu  Wed Jun 12 00:07:12 2013
From: ganger at ece.cmu.edu (Greg Ganger)
Date: Wed, 12 Jun 2013 00:07:12 -0400 (EDT)
Subject: [Disksim-users] A question about time-limited request
In-Reply-To: <CAALB-o8MrZt9G8EBpN7e4req0OLVhpqFvrGHEbLbqeNJnwt2Gw@mail.gmail.com>
References: <CAALB-o8MrZt9G8EBpN7e4req0OLVhpqFvrGHEbLbqeNJnwt2Gw@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.1306120005560.11465@nirvana.pdl.cmu.edu>


No.  If it finishes before the time limit, its generator doesn't block
waiting on it.  Think of it like a prefetch, for example, where the
reader only blocks if the prefetch doesn't finish before the reader
actually needs the data.

Greg


On Tue, 11 Jun 2013, q q wrote:

> If a time-limited request is serviced before its time limit, does the
> generator process block and wait till the end of the time limit before
> generating the next request?
>
> Thank you.
>
> On Page 42 of the Disksim 4.0 manual,
>
> Probability of time-limited request
> "This specifies the probability that a generated request is
> time-limited. That is, the corresponding generator process ?blocks?
> and waits for the request to complete (if it is not already complete)
> after a given amount of think time (specified by the below ?time
> limit? parameters) [5, 2]."
> _______________________________________________
> Disksim-users mailing list
> Disksim-users at ece.cmu.edu
> https://sos.ece.cmu.edu/mailman/listinfo/disksim-users
>

From 6375625 at gmail.com  Wed Jun 12 00:21:00 2013
From: 6375625 at gmail.com (q q)
Date: Tue, 11 Jun 2013 22:21:00 -0600
Subject: [Disksim-users] A question about time-limited request
In-Reply-To: <Pine.LNX.4.64.1306120005560.11465@nirvana.pdl.cmu.edu>
References: <CAALB-o8MrZt9G8EBpN7e4req0OLVhpqFvrGHEbLbqeNJnwt2Gw@mail.gmail.com>
	<Pine.LNX.4.64.1306120005560.11465@nirvana.pdl.cmu.edu>
Message-ID: <CAALB-o_YdVvdkCi1pbksBNY5JxW3g1N4WipO1bozPFDJB2m52Q@mail.gmail.com>

Dear Dr Ganger,

Thank you very much for your prompt response.

This is what I want to do with the synthetic workload generator.
There are two generators - A and B. Both of them generate
time-critical requests with probability of 1.0. Generator A generates
the next request immediately once a request is serviced. Generator B
waits for 2000 ms to generate the next request after a request is
serviced. Can I do this using the synthetic workload generator? If
yes, how?

BTW, I don't think I can do it with the "Think time from request to
return" option of disksim_synthio because that is a global option for
all generators.

Looking forward to hearing from you. Thank you very much

On Tue, Jun 11, 2013 at 10:07 PM, Greg Ganger <ganger at ece.cmu.edu> wrote:
>
>
> No.  If it finishes before the time limit, its generator doesn't block
> waiting on it.  Think of it like a prefetch, for example, where the
> reader only blocks if the prefetch doesn't finish before the reader
> actually needs the data.
>
> Greg
>
>
>
> On Tue, 11 Jun 2013, q q wrote:
>
>> If a time-limited request is serviced before its time limit, does the
>> generator process block and wait till the end of the time limit before
>> generating the next request?
>>
>> Thank you.
>>
>> On Page 42 of the Disksim 4.0 manual,
>>
>> Probability of time-limited request
>> "This specifies the probability that a generated request is
>> time-limited. That is, the corresponding generator process ?blocks?
>> and waits for the request to complete (if it is not already complete)
>> after a given amount of think time (specified by the below ?time
>> limit? parameters) [5, 2]."
>> _______________________________________________
>> Disksim-users mailing list
>> Disksim-users at ece.cmu.edu
>> https://sos.ece.cmu.edu/mailman/listinfo/disksim-users


From ganger at ece.cmu.edu  Wed Jun 12 13:52:37 2013
From: ganger at ece.cmu.edu (Greg Ganger)
Date: Wed, 12 Jun 2013 13:52:37 -0400 (EDT)
Subject: [Disksim-users] A question about time-limited request
In-Reply-To: <CAALB-o_YdVvdkCi1pbksBNY5JxW3g1N4WipO1bozPFDJB2m52Q@mail.gmail.com>
References: <CAALB-o8MrZt9G8EBpN7e4req0OLVhpqFvrGHEbLbqeNJnwt2Gw@mail.gmail.com>
	<Pine.LNX.4.64.1306120005560.11465@nirvana.pdl.cmu.edu>
	<CAALB-o_YdVvdkCi1pbksBNY5JxW3g1N4WipO1bozPFDJB2m52Q@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.1306121351570.21121@nirvana.pdl.cmu.edu>


Yes, I think you should be able to do that, using the parameters as
described in the guide.  Good luck!

Greg


On Tue, 11 Jun 2013, q q wrote:

> Dear Dr Ganger,
>
> Thank you very much for your prompt response.
>
> This is what I want to do with the synthetic workload generator.
> There are two generators - A and B. Both of them generate
> time-critical requests with probability of 1.0. Generator A generates
> the next request immediately once a request is serviced. Generator B
> waits for 2000 ms to generate the next request after a request is
> serviced. Can I do this using the synthetic workload generator? If
> yes, how?
>
> BTW, I don't think I can do it with the "Think time from request to
> return" option of disksim_synthio because that is a global option for
> all generators.
>
> Looking forward to hearing from you. Thank you very much
>
> On Tue, Jun 11, 2013 at 10:07 PM, Greg Ganger <ganger at ece.cmu.edu> wrote:
>>
>>
>> No.  If it finishes before the time limit, its generator doesn't block
>> waiting on it.  Think of it like a prefetch, for example, where the
>> reader only blocks if the prefetch doesn't finish before the reader
>> actually needs the data.
>>
>> Greg
>>
>>
>>
>> On Tue, 11 Jun 2013, q q wrote:
>>
>>> If a time-limited request is serviced before its time limit, does the
>>> generator process block and wait till the end of the time limit before
>>> generating the next request?
>>>
>>> Thank you.
>>>
>>> On Page 42 of the Disksim 4.0 manual,
>>>
>>> Probability of time-limited request
>>> "This specifies the probability that a generated request is
>>> time-limited. That is, the corresponding generator process ?blocks?
>>> and waits for the request to complete (if it is not already complete)
>>> after a given amount of think time (specified by the below ?time
>>> limit? parameters) [5, 2]."
>>> _______________________________________________
>>> Disksim-users mailing list
>>> Disksim-users at ece.cmu.edu
>>> https://sos.ece.cmu.edu/mailman/listinfo/disksim-users
>
>

From mujtaba.tarihi at gmail.com  Mon Jun 17 03:39:58 2013
From: mujtaba.tarihi at gmail.com (Mujtaba Tarihi)
Date: Mon, 17 Jun 2013 12:09:58 +0430
Subject: [Disksim-users] Serious problems using dixtrac extracted parameters
Message-ID: <CA+5nn1gJOPwHxM0COhN5Fjtq4-Bs6jzkkG2UVREKWDZN9Eq6NQ@mail.gmail.com>

Dear all,

I have been attempting to make use of some parameters extracted via
dixtrac, such as the ones graciously provided by Anjo Vahldiek.
However, I have come across very surprising results, performance-wise.
I generated a random trace of aligned 16KB read requests and a
sequential trace of 16KB back-to-back read requests.
This is how long it took to run each trace (it is in milliseconds)
For the record, are the results obtained with the 2GB HP C2490A which
is supplied along with Disksim:
Random: 2232995.378781
Sequential: 624599.551791

To make the comparison (somewhat) fair, I had to use the same random
trace for the larger drives:

And these are the results obtained with the 300GB Maxtor drive:
Random: 237932.073905
Sequential: 619900.348468

And these are the results obtained with the 146GB Maxtor drive:
Random: 345077.024293
Sequential: 329766.817393

And these results are from the 146GB Seagate drive:
Random: 209403.240946
Sequential: 423689.049873

Basically, with the exception of *slight* difference in case of the
146GB maxtor file, the sequential trace takes longer to run, nearly
2.5x times in the worst case!

I just copied the files in the tar.gz files and used them and
layout.mappings is not necessary.

My guess is that the model generated by the tool has issues with the
modelling firmware behavior, something like the scheduling algorithm,
or maybe it is botching up the mappings?

Any help with be more than welcome :)

Regards,
-Tarihi


From vahldiek at mpi-sws.org  Mon Jun 17 06:13:29 2013
From: vahldiek at mpi-sws.org (Anjo Vahldiek)
Date: Mon, 17 Jun 2013 12:13:29 +0200
Subject: [Disksim-users] Serious problems using dixtrac extracted
	parameters
In-Reply-To: <CA+5nn1gJOPwHxM0COhN5Fjtq4-Bs6jzkkG2UVREKWDZN9Eq6NQ@mail.gmail.com>
References: <CA+5nn1gJOPwHxM0COhN5Fjtq4-Bs6jzkkG2UVREKWDZN9Eq6NQ@mail.gmail.com>
Message-ID: <51BEE149.4030009@mpi-sws.org>

Hi,

In general it is difficult to reason about differences without knowing 
how you issued the requests to disks (e.g. O_direct, by SCSI cmd, 
flushing caches, ...) and which exact models you used.

My suggest is that you look at the validation step of dixtrac. At the 
end of dixtrac it runs a validation workload and produces 4 pdfs 
showing the difference between simulated & real device. (called 
mixed(.hist).pdf and random(.hist).pdf which are part of the tar) 
Especially the mixed workload in case of 146g shows a significant 
difference. The 300g model shows less differences.  (not sure which of 
the two you used) You should be able to use the code part to prodcude 
the same pdfs using your disks instead of ours.

Hopefully this helps.

Thanks,
Anjo

On Mon 17 Jun 2013 09:39:58 AM CEST, Mujtaba Tarihi wrote:
> Dear all,
>
> I have been attempting to make use of some parameters extracted via
> dixtrac, such as the ones graciously provided by Anjo Vahldiek.
> However, I have come across very surprising results, performance-wise.
> I generated a random trace of aligned 16KB read requests and a
> sequential trace of 16KB back-to-back read requests.
> This is how long it took to run each trace (it is in milliseconds)
> For the record, are the results obtained with the 2GB HP C2490A which
> is supplied along with Disksim:
> Random: 2232995.378781
> Sequential: 624599.551791
>
> To make the comparison (somewhat) fair, I had to use the same random
> trace for the larger drives:
>
> And these are the results obtained with the 300GB Maxtor drive:
> Random: 237932.073905
> Sequential: 619900.348468
>
> And these are the results obtained with the 146GB Maxtor drive:
> Random: 345077.024293
> Sequential: 329766.817393
>
> And these results are from the 146GB Seagate drive:
> Random: 209403.240946
> Sequential: 423689.049873
>
> Basically, with the exception of *slight* difference in case of the
> 146GB maxtor file, the sequential trace takes longer to run, nearly
> 2.5x times in the worst case!
>
> I just copied the files in the tar.gz files and used them and
> layout.mappings is not necessary.
>
> My guess is that the model generated by the tool has issues with the
> modelling firmware behavior, something like the scheduling algorithm,
> or maybe it is botching up the mappings?
>
> Any help with be more than welcome :)
>
> Regards,
> -Tarihi
> _______________________________________________
> Disksim-users mailing list
> Disksim-users at ece.cmu.edu
> https://sos.ece.cmu.edu/mailman/listinfo/disksim-users


From 815823554 at qq.com  Mon Jun 17 09:48:40 2013
From: 815823554 at qq.com (Ryan)
Date: Mon, 17 Jun 2013 21:48:40 +0800
Subject: [Disksim-users] what should I do with this ?
Message-ID: <000001ce6b61$75638820$602a9860$@com>

Dear all,

 

I?m a new user of disksim . I used disksim-3.0 to test Financial1 trace. I
made some modifies of the related files? disksim_global.h and
disksim_iotrace.c ?in the?src? folder as the disksim reference manual
said to support this new trace format .
However,  it came the following problem:
Wrong number of arguments for I/O trace event type
line: 0,303567,3584,w,0.000000
*** assertion failed: in iotrace_financial1_get_ioreq_event()
(disksim_iotrace.c:573): 0:
Aborted
As a freshman here, I don?t know what to do to deal with this situation .
Would anyone do me a favor?
 
Regards,
-Ryan

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.andrew.cmu.edu/pipermail/disksim-users/attachments/20130617/2309dade/attachment.html>