[gmx-users] RE:RE: One question about Domain decomposition

Berk Hess gmx3 at hotmail.com
Tue Feb 17 09:54:20 CET 2009


Hi,

I would help a lot if you would tell us what you are trying to accomplish.

When charge groups move from one node to an other, only
the global charge group index is communicated (and of course
x and v of the atoms in the cg). Since every node knows
the global topology it can figure out all the properties of the atoms
in for the new cg.

Berk

> Date: Tue, 17 Feb 2009 15:19:47 +0800
> From: xuji at home.ipe.ac.cn
> To: gmx-users at gromacs.org
> Subject: [gmx-users] RE:RE: One question about Domain decomposition
> 
> Hi 
> 
> Thank you for your reply, Berk. I mean that every node has 
> the topology of the whole system. Originally I thought every 
> node has the positions and other properties(information in my 
> last e-mail) of the atoms. I made a mistake obviously. 
> But now, I don't know how do you store the information of the 
> atoms. How do you orgnize the local home atoms and atoms 
> which communicated from neighbour nodes? I think the charge 
> groups' indices are not continuous in one local node, so I
> cann't understand the way you store the atoms' information.
> Can you explain to me in detail? Or give me a sketch map of
> your method? 
> 
> Thank you!
> 	
> Best
> 	wishes!
>  				
> 
>         xuji
>         xuji at home.ipe.ac.cn
>           2009-02-17

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20090217/de45fa01/attachment.html>


More information about the gromacs.org_gmx-users mailing list