Differences

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

--- cluster:73 [2021/03/01 17:09]
hmeij07 [Rstudio]
+++ cluster:73 [2021/06/10 12:55]
hmeij07 [Ovito]
@@ Line 18: / Line 18: @@
 on either cottontail2 or node n33 located at /share/apps/CENTOS6 \\
 Some compilations require 7.x and are in /share/apps/CENTOS7 \\
+====== pytraj ======
+  * https://amber-md.github.io/pytraj/latest/installation.html#installation
+  * From AMBER distribution (Linux, OSX)
+  * pytraj is included in AMBER (version >= 16): ambermd.org
+  * so below I opted for source code installation with python 3.8.3
+  * git clone https://github.com/Amber-MD/pytraj; cd pytra
+  * python ./setup.py install
+  * requires CentOS7 so queues mwgpu, amber128 and exx96
+<code>
+ export PATH=/share/apps/CENTOS7/python/3.8.3/bin:$PATH
+ export LD_LIBRARY_PATH=/share/apps/CENTOS7/python/3.8.3/lib:$LD_LIBRARY_PATH
+# An environment resource file for CPPTRAJ has been created:
+  /zfshomes/apps/CENTOS7/amber/pytraj/cpptraj/cpptraj.sh
+# You may 'source' this file to set up your environment for CPPTRAJ.
+Installed /zfshomes/apps/CENTOS7/python/3.8.3/lib/python3.8/site-packages/pytraj-2.0.6.dev0-py3.8-linux-x86_64.egg
+Processing dependencies for pytraj==2.0.6.dev0
+Searching for numpy==1.20.3
+Best match: numpy 1.20.3
+Adding numpy 1.20.3 to easy-install.pth file
+Installing f2py script to /share/apps/CENTOS7/python/3.8.3/bin
+Installing f2py3 script to /share/apps/CENTOS7/python/3.8.3/bin
+Installing f2py3.8 script to /share/apps/CENTOS7/python/3.8.3/bin
+Using /zfshomes/apps/CENTOS7/python/3.8.3/lib/python3.8/site-packages
+Finished processing dependencies for pytraj==2.0.6.dev0
+</code>
+====== Ovito ======
+  * https://www.ovito.org
+  * /usr/local/bin/ovito -> /usr/local/ovito-basic-3.5.0-x86_64/bin/ovito
+  * on nodes ''cottontail'' and ''greentail52''
+  * uses freeglut/freeglut-devel libraries for opengl
+====== OpenHPC ======
+So I remember where I put the repos ''/share/apps/src/greentail52/openhpc''
+  * whitetail tar ball for centos7/warewulf (no slurm)
+  * http://repos.openhpc.community/OpenHPC/2.0RC1/CentOS_8/x86_64/
+  * http://repos.openhpc.community/ohpc-1.3/1.3.9/base/CentOS_7/x86_64/
+  * https://github.com/warewulf/warewulf3 (master git zip file)
+====== OneAPI ======
+  * Requires CentOS 7 so queues mwgpu, amber128, exx96
+  * Production version is installed in /share/apps/CENTOS7/intel/openapi (via **n78**)
+  * source /share/apps/CENTOS/intel/oneapi/setvars.sh
+<code>
+# which icc icx icpc icpx dpcpp ifort ifx mpirun mpicc python clang clang++
+/share/apps/CENTOS7/intel/oneapi/compiler/2021.1.1/linux/bin/intel64/icc
+/share/apps/CENTOS7/intel/oneapi/compiler/2021.1.1/linux/bin/icx
+/share/apps/CENTOS7/intel/oneapi/compiler/2021.1.1/linux/bin/intel64/icpc
+/share/apps/CENTOS7/intel/oneapi/compiler/2021.1.1/linux/bin/icpx
+/share/apps/CENTOS7/intel/oneapi/compiler/2021.1.1/linux/bin/dpcpp
+/share/apps/CENTOS7/intel/oneapi/compiler/2021.1.1/linux/bin/intel64/ifort
+/share/apps/CENTOS7/intel/oneapi/compiler/2021.1.1/linux/bin/ifx
+/share/apps/CENTOS7/intel/oneapi/intelpython/latest/bin/mpirun
+/share/apps/CENTOS7/intel/oneapi/mpi/2021.1.1/bin/mpicc
+/share/apps/CENTOS7/intel/oneapi/intelpython/latest/bin/python
+/share/apps/CENTOS7/intel/oneapi/compiler/2021.1.1/linux/bin/clang
+/share/apps/CENTOS7/intel/oneapi/compiler/2021.1.1/linux/bin/clang++
+# in addition these were installed
+/share/apps/CENTOS7/intel/oneapi/mkl/2021.1.1
+/share/apps/CENTOS7/intel/oneapi/pytorch/1.5.0
+/share/apps/CENTOS7/intel/oneapi/tensorflow/2.2.0
+</code>
+  * Test version is installed on local disk on host **greentail52**
+  * source /opt/intel/oneapi/setvars.sh
+  * Read [[cluster:203|ICC vs ICX]]
+====== StringTie2 ======
+  * Requires CentOS 7 so queues mwgpu, amber128, exx96
+  * /share/apps/CENTOS7/stringtie2/stringtie2 binarie
+  * https://github.com/skovaka/stringtie2
 ====== Rstudio ======
@@ Line 1093: / Line 1184: @@
 [[https://dokuwiki.wesleyan.edu/doku.php?id=cluster:119|GPU version]]
-  * program: mcc, math, mathematica (v 12.1)
+  * program: mcc, math, mathematica (v 12.2)
-  * install dir: /share/apps/CENTOS6/mathematica/12.1
+  * install dir: /share/apps/CENTOS7/mathematica/12.2
   * docs: on the web
   * www:  http://www.wolfram.com/mathematica/
-  * Campus wide license, use any queue.
+  * Centos7, use queues mwgpu, amber128 or exx96
+  * For GUI launch on greentail52
   * Parallel kernels
@@ Line 1288: / Line 1380: @@
 ====== LAMMPS CPU/GPU ======
+** USER_QUIP **
+This is a version of lammps (29Oct20, last stable) that contains all packages with user-quip.
+It does compile with MPI (mpif90), not ifort, in CentOS7 ... was tricky, solution for MPI was the ''mpifort --showme:link'' answer
+  * Requires CentOS 7 so queues: mwgpu, amber128, exx96 for cpu-only binaries
+  * Rwequires CentOS7 plus Cuda 10.2 so queue: exx96 for cpu+gpu binaries
+Read, if you wish
+  * /share/apps/CENTOS7/lammps/QUIP-public.install
+  * /share/apps/CENTOS7/lammps/29Oct2020.install
+**Environments**
+<code>
+# serial - cpu only (although quippy compiled, it refused to install)
+# MPI STUBS is compiled in
+export PATH=/share/apps/CENTOS7/python/3.8.3/bin:$PATH
+export LD_LIBRARY_PATH=/share/apps/CENTOS7/python/3.8.3/lib:$LD_LIBRARY_PATH
+/share/apps/CENTOS7/lammps/29Oct2020/lmp_serial-with-quip.cpu -h
+# mpi - cpu only
+# gfortran_openmpi
+export QUIP_LIBRARY=/share/apps/CENTOS7/lammps/QUIP-public/9Apr2021/linux_x86_64_gfortran_openmpi
+export PATH=/share/apps/CENTOS7/python/3.8.3/bin:$PATH
+export LD_LIBRARY_PATH=/share/apps/CENTOS7/python/3.8.3/lib:$LD_LIBRARY_PATH
+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
+/share/apps/CENTOS7/lammps/29Oct2020/lmp_mpi-with-quip.cpu -h
+# mpi - cpu+gpu
+# gfortran_openmpi arch
+export PATH=/share/apps/CENTOS7/python/3.8.3/bin:$PATH
+export LD_LIBRARY_PATH=/share/apps/CENTOS7/python/3.8.3/lib:$LD_LIBRARY_PATH
+export QUIP_LIBRARY=/share/apps/CENTOS7/lammps/QUIP-public/9Apr2021/linux_x86_64_gfortran_openmpi
+export PATH=/share/apps/CENTOS7/python/3.8.3/bin:$PATH
+export LD_LIBRARY_PATH=/share/apps/CENTOS7/python/3.8.3/lib:$LD_LIBRARY_PATH
+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
+export CUDA_HOME=/usr/local/cuda
+export PATH=/usr/local/cuda/bin:$PATH
+export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
+export LD_LIBRARY_PATH=/usr/local/cuda/lib:$LD_LIBRARY_PATH
+# the "make" versions of lmp with quip do not work, deleted them
+# did find on nvdia dev forum reference problem disappears when using cmake (vs make)
+# these lmp binaries do work
+/share/apps/CENTOS7/lammps/29Oct2020/single_single/bin/lmp
+/share/apps/CENTOS7/lammps/29Oct2020/single_double/bin/lmp
+/share/apps/CENTOS7/lammps/29Oct2020/double_double/bin/lmp
+# however I had to load non-mpi version of libquid.a into lmp
+# witch seems ok given parallel work is driven by -suffix gpu
+# sample script: ~hmeij/k20redo/run.rtx.lammps2
+</code>
 ** CPU only ** -Centos6
@@ Line 3003: / Line 3162: @@
 ===== Intel  =====
-UPDATE, 2016 parallel studio cluster edition installed
+UPDATE 2021
+latest version of icc/icx and ifort/ifx are inside OneAPI, search this page
+UPDATE 2016
+parallel studio cluster edition installed
   * 2 ifort and 2 icc float licenses
@@ Line 3031: / Line 3194: @@
 ====== Cluster Math Kernel Libraries ======
+  * latest version in OneAPI, search this page
   * see Intel parallel studio
@@ Line 3040: / Line 3205: @@
 ====== Intel Threading Building Blocks ======
+  * latest version is in OneAPI, search this pages
   * see Intel parallel studio
@@ Line 3054: / Line 3221: @@
   * program: Matlab, link to binary is 'matlab' in ''/share/apps/bin''
-    * R2020a, campus wide license so no more restrictions on licenses
+    * R2020b, campus wide license so no more restrictions on licenses
     * No Distributed Computing Engine or Matlab Parallel (for now)
-  * install dir: /share/apps/CENTOS7/matlab/R2020a
+  * install dir: /share/apps/CENTOS7/matlab/R2020bISO
   * docs & www: http://mathworks.com
+  * CentOS7 required use on queues mwgpu, amber128 and exx96
+  * For GUI launch on greentail52
 Note:
@@ Line 3071: / Line 3239: @@
 ====== Stata ======
-  * program: Stata v 15, 6-user network license, includes stata, stata-mp, stata-se
+  * program: Stata v 17, 6-user network license, includes stata, stata-mp, stata-se
-  * install dir: /share/apps/stata/15
+  * install dir: /share/apps/stata/
   * docs: at web site
   * http://stata.com
 <code>
-export PATH=/share/apps/stata/15:$PATH
+export PATH=/share/apps/stata/17:$PATH
 </code>

DokuWiki

User Tools

Site Tools

Differences

Page Tools