Re: [petsc-users] An issue about pipelined CG and Gropp's CG

[赵刚]

Dear Barry,

Thank you for your prompt reply. I run ~16M DOFs on 32 nodes (36 cores per 
node), but CG seems to be faster than pipelined CG and Gropp's CG, I'm puzzled 
and haven't figured out why. Put the performance output into attachment, please 
check it.



Thanks,
Gang


> -----原始邮件-----
&gt; 发件人: "Barry Smith" <bsm...@petsc.dev>
&gt; 发送时间: 2021-02-18 09:17:17 (星期四)
&gt; 收件人: "赵刚" <zha...@lsec.cc.ac.cn>
&gt; 抄送: PETSc <petsc-users@mcs.anl.gov>
&gt; 主题: Re: [petsc-users] An issue about pipelined CG and Gropp's CG
&gt; 
&gt; 
&gt; 
&gt; &gt; On Feb 17, 2021, at 6:47 PM, 赵刚 <zha...@lsec.cc.ac.cn> wrote:
&gt; &gt; 
&gt; &gt; Dear PETSc team,
&gt; &gt; 
&gt; &gt; I am interested in pipelined CG (-ksp_type pipecg) and Gropp's CG 
(-ksp_type groppcg), it is expected that this iterative method with pipelined 
has advantages over traditional CG in the case of multiple processes. So I'd 
like to ask for Poisson problem, how many computing nodes do I need to show the 
advantages of pipelined CG or Gropp's CG over CG (No preconditioner is used)?
&gt; &gt; 
&gt; &gt; Currently, I can only use up to 32 nodes (36 cores per nodes) at most 
on my cluster, but both "pipecg" and "groppcg" seem to be no advantage over 
"cg" when I solve Poisson equations with homogeneous Dirichlet BC in [0, 1]^2 
(remain 20K~60K DOFs per process). I guess the reason would be too few 
computing nodes.
&gt; 
&gt;   900 cores (assuming they are not memory bandwidth bound) might be enough 
to see some differences but the differences are likely so small compared to 
other parallel issues that affect performance that you see no consistently 
measurable difference.
&gt; 
&gt;    Run with -log_view three cases, no pipeline and the two pipelines and 
send the output. By studying where the time is spent in the different regions 
of the code with this output one may be able to say something about the 
pipeline affect.
&gt; 
&gt;   Barry
&gt; 
&gt; 
&gt; &gt; 
&gt; &gt; Because I am calling PETSc via other numerical software, if need, I 
would mail related performance information to you by using command line options 
suggested by PETSc. Thank you.
&gt; &gt; 
&gt; &gt; 
&gt; &gt; Thanks,
&gt; &gt; Gang
</zha...@lsec.cc.ac.cn></petsc-users@mcs.anl.gov></zha...@lsec.cc.ac.cn></bsm...@petsc.dev>

cg.out
Description: Binary data

groppcg.out
Description: Binary data

pipecg.out
Description: Binary data