My guys ran last night without any problem. However, I want to double check
with Roy here because I'm not convinced it actually ran multithreaded. More
to come later today...

On Thu, Oct 20, 2011 at 8:31 AM, David Fuentes <[email protected]> wrote:

> wow! thanks!
>
> I can create a simple regression test for a diffusion equation if that will
> be useful
>
>
> df
>
>
>
> On Wed, Oct 19, 2011 at 6:01 PM, Roy Stogner <[email protected]>wrote:
>
>>
>> On Wed, 19 Oct 2011, Roy Stogner wrote:
>>
>>  so why the hell haven't I added TBB to FEMSystem yet?"
>>>
>>
>> FEMSystem is now multithreaded (for assembly, postprocessing, qoi
>> assembly, and qoi derivative assembly) as of r4878.  It's passing my
>> basic initial checks; we'll hit it with some stricter regression tests
>> shortly.
>>
>> If you're experimenting, though, note the tradeoff here: PETSc is
>> *not* multithreaded, so --n_threads greater than 1 is really mostly of
>> value for problems where you're not performance-bound by the algebraic
>> solve (or where you can make the heavy part of the solve multithreaded
>> - I think PETSc integrates with some multithreaded 3rd party
>> preconditioners?)
>> ---
>> Roy
>>
>
>
------------------------------------------------------------------------------
The demand for IT networking professionals continues to grow, and the
demand for specialized networking skills is growing even more rapidly.
Take a complimentary Learning@Ciosco Self-Assessment and learn 
about Cisco certifications, training, and career opportunities. 
http://p.sf.net/sfu/cisco-dev2dev
_______________________________________________
Libmesh-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/libmesh-users

Reply via email to