Add versioninfo to MSE comparsion

[Sbozzolo]

Not a solution to unstable MSE problems, but this will print something like:

Julia Version 1.9.3
Commit bed2cd540a1 (2023-08-24 14:43 UTC)
Build Info:
  Official https://julialang.org/ release
Platform Info:
  OS: Linux (x86_64-linux-gnu)
  CPU: 32 × Intel(R) Xeon(R) Gold 6130 CPU @ 2.10GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-14.0.6 (ORCJIT, skylake-avx512)
  Threads: 1 on 32 virtual cores
Environment:
  LD_LIBRARY_PATH = /central/software/hdf5/1.12.2-ompi415//lib:/central/software/OpenMPI/4.1.5_cuda-12.2//lib:/central/software/ucx/1.14.1_cuda-12.2//lib:/central/software/CUDA/12.2/lib64:/central/software/CUDA/12.2/extras/CUPTI/lib64:/central/software/CUDA/12.2/targets/x86_64-linux/lib:/central/software/julia/1.9.3/lib
  LD_RUN_PATH = /central/software/OpenMPI/4.1.5_cuda-12.2//lib:/central/software/ucx/1.14.1_cuda-12.2//lib:/central/software/CUDA/12.2/lib64:/central/software/CUDA/12.2/extras/CUPTI/lib64:/central/software/CUDA/12.2/targets/x86_64-linux/lib

This can help identifying runs that are known to have MSE problems (e.g., because they are not broadwell nodes).

[simonbyrne]

If you just want the features, we can also get that by

scontrol show node $SLURM_NODELIST

I can add this to the default output

[simonbyrne]

I've added this to the environment script, so now it should now it should display something like:

NodeName=hpc-92-37 Arch=x86_64 CoresPerSocket=14
   CPUAlloc=22 CPUEfctv=28 CPUTot=28 CPULoad=20.66
   AvailableFeatures=broadwell
   ActiveFeatures=broadwell
   Gres=gpu:p100:4
   NodeAddr=hpc-92-37 NodeHostName=hpc-92-37 Version=22.05.6
   OS=Linux 3.10.0-1160.53.1.el7.x86_64 #1 SMP Fri Jan 14 13:59:45 UTC 2022
   RealMemory=250000 AllocMem=145408 FreeMem=187328 Sockets=2 Boards=1
   State=MIXED+RESERVED ThreadsPerCore=1 TmpDisk=0 Weight=30 Owner=N/A MCS_label=N/A
   Partitions=any
   BootTime=2023-09-20T09:28:59 SlurmdStartTime=2023-11-02T07:05:01
   LastBusyTime=2023-11-09T09:28:17
   CfgTRES=cpu=28,mem=250000M,billing=28,gres/gpu=4
   AllocTRES=cpu=22,mem=142G,gres/gpu=2
   CapWatts=n/a
   CurrentWatts=0 AveWatts=0
   ExtSensorsJoules=n/s ExtSensorsWatts=0 ExtSensorsTemp=n/s

[Sbozzolo]

If you just want the features, we can also get that by

scontrol show node $SLURM_NODELIST

I can add this to the default output

Yes, but I wanted to have everything immediately available on the job.