Skip to Content.
Sympa Menu

charm - Re: [charm] [ppl] Projections

charm AT lists.cs.illinois.edu

Subject: Charm++ parallel programming system

List archive

Re: [charm] [ppl] Projections


Chronological Thread 
  • From: Yanhua Sun <sun51 AT illinois.edu>
  • To: "Kale, Laxmikant V" <kale AT illinois.edu>
  • Cc: "charm AT cs.uiuc.edu" <charm AT cs.uiuc.edu>
  • Subject: Re: [charm] [ppl] Projections
  • Date: Sat, 26 Nov 2011 12:44:51 -0600
  • List-archive: <http://lists.cs.uiuc.edu/pipermail/charm>
  • List-id: CHARM parallel programming system <charm.cs.uiuc.edu>


Hi Tauro,
Can you try to delete all the .sum files in your directory and then run projections again?  I did also meet projections problems with displaying .sum files. We will take a look.  On the other hand, mostly sum files are unnecessary.   Since all the information .sum files provide is also included in .log files, we usually do not generate .sum .log files together.  To not generate sum files, you can edit Makefile in NAMD and search 'tracemode summary' and remove it. 

Yanhua

On Fri, Nov 25, 2011 at 3:44 AM, Kale, Laxmikant V <kale AT illinois.edu> wrote:
Reinstall Charm++ without the  "with-production" flag. . It turns off instrumentation completely, according to the Projections manual.

(You can have both versions of charm++ around, and just link with the with-production version when you are not doing projections analysis)

Sanjay
-- 
Laxmikant (Sanjay) Kale         http://charm.cs.uiuc.edu
Professor, Computer Science     kale AT illinois.edu
201 N. Goodwin Avenue           Ph:  (217) 244-0094
Urbana, IL  61801-2302          FAX: (217) 265-6582

On 11/25/11 2:56 AM, "takuro" <udagawa.t.aa AT m.titech.ac.jp> wrote:

Hi,

I'd like to use projections and analyze the performance of NAMD2.8.
After runnning NAMD and tried to use projection, I got following error and projections was closed.
>../projections/bin/projections namd2.prj.sts
Found 11 log files
Found 11 sum files
[1] Token read = [0+2147483647] is impossible as the count is larger than the total count of 35106

I downloaded projections from the git repository and ran on my desktop PC running ubuntu 10.10.
Could anyone give advice about this problem?

Building and compiling procedure of Charm++ and NAMD was below.
> cd NAMD_2.8/charm-6.3.2
> ./build charm++ net-linux-x86_64 gcc ibverbs smp -j16 --with-production --enable-tracing
> cd ..
>./config Linux-x86_64-g++ --charm-arch net-linux-x86_64-ibverbs-smp-gcc
> cd Linux-x86_64-g++
> make
> make projections
> make summary

The scripts and NAMD configuration file were bellow

throw_job.sh:
#!/bin/sh

~/NAMD_2.8/charm-6.3.2/net-linux-x86_64-ibverbs-smp-gcc/bin/charmrun +p11 ++ppn 11 ++mpiexec ++remote-shell ~/mympiexec/mympiexec.sh ++verbose ~/NAMD_2.8/Linux-x86_64-g++/namd2.prj +logsize 10000000 +traceroot prj ~/NAMD_projection/apoa1.namd > output.log

mympiexec.sh:
exec mpiexec -hostfile hostfile.dat $*

apoa1.namd:
cellBasisVector1     108.8612 0.0 0.0
cellBasisVector2     0.0 108.8612 0.0
cellBasisVector3     0.0 0.0 77.758
cellOrigin           0.0 0.0 0.0

coordinates          apoa1.pdb
temperature          300
seed                 74269

switching            on
switchdist           10
cutoff               12
pairlistdist         13.5
margin               0
stepspercycle        20

PME                  on
PMEGridSizeX         108
PMEGridSizeY         108
PMEGridSizeZ         80

structure            apoa1.psf
parameters           par_all22_prot_lipid.xplor
parameters           par_all22_popc.xplor
exclude              scaled1-4
1-4scaling           1.0

timestep             1.0
fullElectFrequency   4

numsteps             500
outputtiming         100

outputname           /usr/tmp/apoa1-out

Thanks,

Takuro Udagawa


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

_______________________________________________
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




--
Yours Sincerely,
Yanhua Sun (Jessie)
PhD student
Parallel Programming Lab
Department of Computer Science,
University of Illinois at Urbana-Champaign,
201 N. Goodwin Ave.
Urbana, IL 61801




Archive powered by MHonArc 2.6.16.

Top of Page