Differences

This shows you the differences between two versions of the page.

--- cluster:207 [2021/10/11 17:11]
hmeij07 [Slurm Test]
+++ cluster:207 [2023/10/27 18:47] (current)
hmeij07
@@ Line 2: / Line 2: @@
 **[[cluster:0|Back]]**
+**Make sure munge/unmunge work between 1.3/2.4, that date is in sync (else you get error #16)**
 ===== Slurm Test Env =====
@@ Line 7: / Line 8: @@
 Getting a head start on our new login node plus two cpu+gpu compute node project. Hardware has been purchased but there is long delivery time. Meanwhile it makes sense to setup a standalone Slurm scheduler and do some testing and have as a backup. Slurm will be running on ''greentail52'' with a some compute nodes.
+This page just intended to keep documentation sources handy. Go to the **Users** page [[cluster:208|Slurm Test Env]]
-**SLURM documentation**
+==== SLURM documentation ====
 <code>
@@ Line 33: / Line 35: @@
 https://slurm.schedmd.com/slurm.conf.html
 section: node configuration
+The node range expression can contain one pair of square brackets with a sequence of comma-separated numbers and/or ranges of numbers separated by a "-" (e.g. "linux[0-64,128]", or "lx[15,18,32-33]")
 Features (hasGPU, hasRTX5000)
 are intended to be used to filter nodes eligible to run jobs via the --constraint argument.
@@ Line 49: / Line 54: @@
 https://slurm.schedmd.com/gres.html#GPU_Management
 setting up gres.conf
+give GPU jobs priority using the Multifactor Priority plugin:
+https://slurm.schedmd.com/priority_multifactor.html#tres
+PriorityWeightTRES=GRES/gpu=1000
+example here: https://slurm.schedmd.com/SLUG19/Priority_and_Fair_Trees.pdf
+requires faishare thus the database
 https://slurm.schedmd.com/mc_support.html
@@ Line 106: / Line 117: @@
-** MUNGE installation**
+==== MUNGE installation ====
 <code>
@@ Line 172: / Line 183: @@
 </code>
-** SLURM installation **
+==== SLURM installation Updated ====
 <code>
-source /share/apps/CENTOS7/amber/miniconda3/etc/profile.d/conda.sh
+export PATH=/usr/local/cuda/bin:$PATH
-export PATH=/share/apps/CENTOS7/amber/miniconda3/bin:$PATH
+export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
-export LD_LIBRARY_PATH=/share/apps/CENTOS7/amber/miniconda3/lib:$LD_LIBRARY_PATH
-which mpirun python conda
-# cuda 9.2
+[root@cottontail2 slurm-22.05.2]# which gcc mpicc nvcc
-export CUDAHOME=/usr/local/n37-cuda-9.2
+/opt/ohpc/pub/compiler/gcc/9.4.0/bin/gcc
-export PATH=/usr/local/n37-cuda-9.2/bin:$PATH
+/opt/ohpc/pub/mpi/openmpi4-gnu9/4.1.1/bin/mpicc
-export LD_LIBRARY_PATH=/usr/local/n37-cuda-9.2/lib64:$LD_LIBRARY_PATH
+/usr/local/cuda/bin/nvcc
-which nvcc
- export PATH=/share/apps/CENTOS7/openmpi/4.0.4/bin:$PATH
- export LD_LIBRARY_PATH=/share/apps/CENTOS7/openmpi/4.0.4/lib:$LD_LIBRARY_PATH
- which mpirun
+./configure \
+--prefix=/usr/local/slurm-22.05.2 \
+--sysconfdir=/usr/local/slurm-22.05.2/etc \
+--with-nvml=/usr/local/cuda
+make
+make install
+export PATH=/usr/local/slurm/bin:$PATH
+export LD_LIBRARY_PATH=/usr/local/slurm/lib:$LD_LIBRARY_PATH
+[root@cottontail2 slurm-22.05.2]# find /usr/local/slurm-22.05.2/ -name auth_munge.so
+/usr/local/slurm-22.05.2/lib/slurm/auth_munge.so
+</code>
+==== SLURM installation ====
+Configured and compiled on ''greentail52'' despite not having gpus...only library manager is needed (nvml)
+<code>
+# cuda 9.2 ...
+# installer found /usr/local/cuda on ''greentail''
+# just in case
+export PATH=/share/apps/CENTOS7/openmpi/4.0.4/bin:$PATH
+export LD_LIBRARY_PATH=/share/apps/CENTOS7/openmpi/4.0.4/lib:$LD_LIBRARY_PATH
+which mpirun
+# /usr/local/slurm is symbolic link to slurm-21.08.1
 ./configure \
 --prefix=/usr/local/slurm-21.08.1 \
 --sysconfdir=/usr/local/slurm-21.08.1/etc \
---with-nvml=/usr/local/n37-cuda-9.2 \
+ | tee -a install.log
--with-hdf5=no | tee -a install.log
+# skip # --with-nvml=/usr/local/n37-cuda-9.2 \
+# skip # -with-hdf5=no  \
 # known hdf5 library problem when including --with-nvml
@@ Line 203: / Line 238: @@
 config.status: creating src/plugins/gpu/nvml/Makefile
+====
+Libraries have been installed in:
+   /usr/local/slurm-21.08.1/lib/slurm
+If you ever happen to want to link against installed libraries
+in a given directory, LIBDIR, you must either use libtool, and
+specify the full pathname of the library, or use the '-LLIBDIR'
+flag during linking and do at least one of the following:
+   - add LIBDIR to the 'LD_LIBRARY_PATH' environment variable
+     during execution
+   - add LIBDIR to the 'LD_RUN_PATH' environment variable
+     during linking
+   - use the '-Wl,-rpath -Wl,LIBDIR' linker flag
+   - have your system administrator add LIBDIR to '/etc/ld.so.conf'
+====
+# for now
 export PATH=/usr/local/slurm/bin:$PATH
 export LD_LIBRARY_PATH=/usr/local/slurm/lib:$LD_LIBRARY_PATH
-From job completions file, JOB #3
+</code>
+==== General Accounting ====
+<code>
+From job completions file, JOB #3, convert Start and End times to epoch seconds
 StartTime=2021-10-06T14:32:37 EndTime=2021-10-06T14:37:40
@@ Line 222: / Line 280: @@
-**Full Version Slurm Config Tool**
+==== Slurm Config Tool ====
+  * lets start with this file and build up/out
 <code>
@@ Line 372: / Line 432: @@
 #
 # COMPUTE NODES
-NodeName=n[110-111] CPUs=2 RealMemory=192 CoresPerSocket=12 ThreadsPerCore=12 State=UNKNOWN
+NodeName=n[110-111] CPUs=2 RealMemory=192 CoresPerSocket=12 ThreadsPerCore=2 State=UNKNOWN
 #
 #

DokuWiki

User Tools

Site Tools

Differences

Page Tools