[namd3 alpha9] FATAL ERROR: Low global CUDA exclusion count! (781005 vs 781007) System unstable or pairlistdist or cutoff too small.

From: Sungho Han (sungho.han-2_at_postgrad.manchester.ac.uk)
Date: Fri Mar 18 2022 - 11:35:22 CDT

• Next message: Anirvinya Gururajan: "Error while simulating on NAMD"
• Previous message: ROPÓN-PALACIOS G.: "COLVAR moving restraint"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

Hello,

I am simulating protein lipid membrane system with ~560k atoms (PME, NPT). I have been running the identical simulation on CPUs with NAMD2.13, but want to try running the calculations on GPUs.

While running NAMD_3.0alpha9_Linux-x86_64-multicore-CUDA,
namd3 +p1 +idlepoll +setcpuaffinity --CUDASOAintegrate on +devices 0 \
        ${molname}_eq1.namd > ${molname}_eq1.namd.out

I altered following parameters from my successful NAMD2.13 run:
   set margin to 4
   removed "stepsPerCycle" and "pairlistsPerCycle"
   increased pairlistdist to 16 (originally 12.5)

But I get error:

Info: Pairlistdist is too small for 1 computes during timestep 67.

FATAL ERROR: Low global CUDA exclusion count! (781005 vs 781007) System unstable or pairlistdist or cutoff too small.

I read a similar query email back in 2020 (https://www.ks.uiuc.edu/Research/namd/mailing_list/namd-l.2020-2021/0466.html), is there still same problem?

How can I resolve this issue? Thank you in advance for your help.

Best regards,
Bosco

__________________________________

Sungho Bosco Han

A*STAR PhD student
Faculty of Biology, Medicine and Health

The University of Manchester

Oxford Road

Manchester

• Next message: Anirvinya Gururajan: "Error while simulating on NAMD"
• Previous message: ROPÓN-PALACIOS G.: "COLVAR moving restraint"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

This archive was generated by hypermail 2.1.6 : Tue Dec 13 2022 - 14:32:44 CST