Re: [Libmesh-devel] old_dof_object

2008-03-11 Thread Roy Stogner
On Tue, 11 Mar 2008, Benjamin Kirk wrote: >>> It's already too late to save the FE classes, but please, the >>> DofObjects are innocent; there's no need to kill again! >> >> I'm sorry Roy but I agree with Ben, No need to apologize; you're disagreeing with me politely about a technical question.

Re: [Libmesh-devel] old_dof_object

2008-03-11 Thread Benjamin Kirk
>> It's already too late to save the FE classes, but please, the >> DofObjects are innocent; there's no need to kill again! >> > > I'm sorry Roy but I agree with Ben, > I think that the FE classes help making the difference taking libMesh > one step ahead. > But, I have to say that I love templat

Re: [Libmesh-devel] old_dof_object

2008-03-11 Thread Roy Stogner
On Tue, 11 Mar 2008, Benjamin Kirk wrote: > I've got one more easy fix which helps a lot... Combine the _n_vars and > _n_comp char vectors, using _n_comp[s][0] to hold the number of variables in > system s. Then _n_comp[s][1] ... _n_comp[s][NV] is the actual number of > components for each varia

Re: [Libmesh-devel] old_dof_object

2008-03-11 Thread John Peterson
On Tue, Mar 11, 2008 at 6:01 PM, Benjamin Kirk <[EMAIL PROTECTED]> wrote: > >>> Just eliminating that unnecessary, incomplete copy reduced the memory > >>> usage > >>> to 980MB, which I consider a huge savings. So needless to say we will > not > >>> copy the dof objects when they do not con

Re: [Libmesh-devel] old_dof_object

2008-03-11 Thread Benjamin Kirk
>>> Just eliminating that unnecessary, incomplete copy reduced the memory usage >>> to 980MB, which I consider a huge savings. So needless to say we will not >>> copy the dof objects when they do not contain complete information. >> >> Wow! I'd consider that a big savings too. > > I'll third

Re: [Libmesh-devel] old_dof_object

2008-03-11 Thread Roy Stogner
On Tue, 11 Mar 2008, John Peterson wrote: > On Tue, Mar 11, 2008 at 1:10 PM, Benjamin Kirk <[EMAIL PROTECTED]> wrote: >> Just eliminating that unnecessary, incomplete copy reduced the memory usage >> to 980MB, which I consider a huge savings. So needless to say we will not >> copy the dof obje

Re: [Libmesh-devel] old_dof_object

2008-03-11 Thread John Peterson
On Tue, Mar 11, 2008 at 1:10 PM, Benjamin Kirk <[EMAIL PROTECTED]> wrote: > A question recently came up on the user's list regarding the old_dof_object. > Specifically, is it OK to clear it after system projection. > > I couldn't think of a reason why not, but I also figured it had a pretty > sm

[Libmesh-devel] old_dof_object

2008-03-11 Thread Benjamin Kirk
A question recently came up on the user's list regarding the old_dof_object. Specifically, is it OK to clear it after system projection. I couldn't think of a reason why not, but I also figured it had a pretty small memory footprint. Not so, especially on 64-bit machines. I've got a ~1 million e