Well, now I'm using the ResFailure.example01 (it's more simple) and
the issue remains. I understood what you said. What I don't undestand
is when I raise the avg the gridlet takes the same amount of time to
be executed. Changes on the avg seems not to affect the time that
takes to execute the gridlet.
Thanks for you help!
Vinicius Pinheiro
So, the higher the avg for this distribution is, the less
> frequent failures will be
2010/2/23 Agustín Caminero Herráez <ag...@ds...>:
> Hi Vinicius,
>
>
> Vinicius Pinheiro wrote:
>> Hi people,
>>
>> I'm running the ResFailure.example03 and I have some doubts.
>> I run with two failureTimePattern configuration: one with
>> HyperExponential(25, 100, 4) and other with HyperExponential(250, 100,
>> 4);
>>
>> My expectation was that using 250 as avarega instead of 25 would lead
>> to a smalllest CPU time since the interval between failures increased.
>> But the opposite occurred.
>> I don't undestood this behaviour. Can someone give me a help on this?
>>
> From "RegionalGISWithFailure.java", "failureTimePattern: defines on when
> each resource will fail". That is, the moments in time when failures
> will occur. So, the higher the avg for this distribution is, the less
> frequent failures will be.
>
> Regards,
>
> Agustin
>
>
>> Best regards,
>>
>> Vinicius Pinheiro
>>
>> with HyperExponential(25, 100, 4);
>>
>> ------------------------------------------------------------------------------
>> Download Intel® Parallel Studio Eval
>> Try the new software tools for yourself. Speed compiling, find bugs
>> proactively, and fine-tune applications for parallel performance.
>> See why Intel Parallel Studio got high marks during beta.
>> http://p.sf.net/sfu/intel-sw-dev
>> _______________________________________________
>> Gridsim-users mailing list
>> Gri...@li...
>> https://lists.sourceforge.net/lists/listinfo/gridsim-users
>>
>
> --
> Dr. Agustin Caminero
> Assistant Professor
> Dept. of Communication and Control Systems
> The National University of Distance Education
> Madrid. Spain
> E-Mail:acc...@sc...
> Phone: +34 913 989 468
> http://www.uned.es
>
>
|
