[yt-users] clump finder crash

Matthew Turk matthewturk at gmail.com
Wed Jun 13 05:19:18 PDT 2012


On Tue, Jun 12, 2012 at 8:18 PM, Elizabeth Tasker
<tasker at astro1.sci.hokudai.ac.jp> wrote:
> I should maybe add that I'm running this on a disk object and I was previously using connected sets on its own without the clump find wrapper, and that was no problem.
>
> I think that means that either it's the periodic nature that's an issue or it's a memory issue because the number of iterations clump find is doing?

It will output each time it does another iteration, which leads me to
believe something is going wrong the first time.  How have you changed
between the clump and the connected set calls?  Are they operating on
the same data object?

A thought occurs: Stephen, Britton, do you have the binding energy
calculation on by default?

>
> Elizabeth
>
> On 2012-06-13, at 9:09 AM, Matthew Turk wrote:
>
>> Hi Elizabeth,
>>
>> On Tue, Jun 12, 2012 at 7:34 PM, Elizabeth Tasker
>> <tasker at astro1.sci.hokudai.ac.jp> wrote:
>>> Hi,
>>>
>>> I'm having some problems with clump finder.
>>>
>>> Specifically, it gets to this point:
>>>
>>> yt : [INFO     ] 2012-06-13 02:10:34,119 Getting field TotalEnergy from 1
>>> yt : [INFO     ] 2012-06-13 02:10:34,139 Getting field x from 1
>>> yt : [INFO     ] 2012-06-13 02:10:34,147 Getting field y from 1
>>> yt : [INFO     ] 2012-06-13 02:10:34,155 Getting field z from 1
>>> yt : [INFO     ] 2012-06-13 02:10:34,159 Adjusting clump for periodic boundaries in dim z
>>>
>>> and then stops (no error).
>>>
>>> My region actually isn't periodic and this must be a memory intensive step that isn't appropriate. Does the clump find tools/wrapper have a switch for non-periodic regions?
>>
>> Are you sure this is a memory problem?  Does it OOM error, does it
>> segfault, etc etc?
>>
>> As for turning off periodicity, this is possible but not currently implemented.
>>
>> -Matt
>>
>>>
>>> Thanks!
>>>
>>> Elizabeth
>>> _______________________________________________
>>> yt-users mailing list
>>> yt-users at lists.spacepope.org
>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>> _______________________________________________
>> yt-users mailing list
>> yt-users at lists.spacepope.org
>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>
> _______________________________________________
> yt-users mailing list
> yt-users at lists.spacepope.org
> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org



More information about the yt-users mailing list