Skip to Content.
Sympa Menu

charm - Re: [charm] [ppl] Appropriate build type for LLNL linux clusters

charm AT lists.cs.illinois.edu

Subject: Charm++ parallel programming system

List archive

Re: [charm] [ppl] Appropriate build type for LLNL linux clusters


Chronological Thread 
  • From: "Bennion, Brian" <Bennion1 AT llnl.gov>
  • To: Abhinav Bhatele <bhatele AT illinoisalumni.org>
  • Cc: "charm AT cs.uiuc.edu" <charm AT cs.uiuc.edu>
  • Subject: Re: [charm] [ppl] Appropriate build type for LLNL linux clusters
  • Date: Mon, 29 Aug 2011 15:11:03 -0700
  • Accept-language: en-US
  • Acceptlanguage: en-US
  • List-archive: <http://lists.cs.uiuc.edu/pipermail/charm>
  • List-id: CHARM parallel programming system <charm.cs.uiuc.edu>

So far only the vanilla mpi-linux-x86_64 version of charm++ will compile and
pass any tests.

The mpi smp will compile but segfaults when using simplearrayhello.
While the ./build tool recognizes the ibverbs option, there is no directory
for it in the arch/ directory in the charm6.2.1 tarball that I am using.
Brian



-----Original Message-----
From:
bhatele AT gmail.com

[mailto:bhatele AT gmail.com]
On Behalf Of Abhinav Bhatele
Sent: Monday, August 29, 2011 12:01 PM
To: Bennion, Brian
Cc:
charm AT cs.uiuc.edu
Subject: Re: [ppl] [charm] Appropriate build type for LLNL linux clusters

Hi Brian,

You can try a couple different charm builds and then benchmark NAMD to see
which one gives you the best performance:

./build charm++ mpi-linux-x86_64 --with-production
./build charm++ mpi-linux-x86_64 smp --with-production
./build charm++ net-linux-x86_64 ibverbs --with-production
./build charm++ net-linux-x86_64 ibverbs smp --with-production

The smp builds of Charm/NAMD require slightly different runtime options than
the regular builds. Please e-mail us if you have any problems.

- Abhinav



On Mon, Aug 29, 2011 at 11:10 AM, Bennion, Brian
<Bennion1 AT llnl.gov>
wrote:


Hello,

I would like to build the proper version of charm++ for our linux
clusters in order to use NAMD2.8

Based on the specs listed on this website, which build parameters
would give me the best charm++ performance

https://computing.llnl.gov/?set=resources&page=OCF_resources#sierra

In the past, i have built namd/charm++ but the nodes then had only a
handful of cpus, now they are a bit more complicated.

Any suggestions would be helpful.


brian bennion

_______________________________________________
charm mailing list

charm AT cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/charm
_______________________________________________
ppl mailing list

ppl AT cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/ppl





--
Abhinav Bhatele
Postdoctoral Research Staff Member
Center for Applied Scientific Computing
Lawrence Livermore National Laboratory






Archive powered by MHonArc 2.6.16.

Top of Page