Differences

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

--- cluster:73 [2020/06/18 19:26]
hmeij07 [Sequencing Tools #2]
+++ cluster:73 [2020/10/26 19:29]
hmeij07 [LAMMPS CPU/GPU]
@@ Line 3: / Line 3: @@
 ====== Software ======
+IMPORTANT NOTE: **Since moving to the TrueNAS/ZFS appliance all references to** ''/home/apps'' **should be replaced with** ''/share/apps'' **which points to** ''/zfshomes/apps''
+ --- //[[hmeij@wesleyan.edu|Henk]] 2020/07/31 15:08//
 The listings below will be updated as software is installed, location is /share/apps/ or /home/apps/ (same thing).
@@ Line 14: / Line 18: @@
 on either cottontail2 or node n33 located at /share/apps/CENTOS6 \\
 Some compilations require 7.x and are in /share/apps/CENTOS7 \\
+====== EMBOSS ======
+  * /share/apps/CENTOS6/EMBOSS
+  * /share/apps/CENTOS6/EMBOSS/6.6.0.install (read to setup env)
+  * https://www.ebi.ac.uk/services
+  * jfabry/kthayer
 ====== Sequencing Tools #2 ======
@@ Line 22: / Line 33: @@
   * https://broadinstitute.github.io/picard/
   * https://bedtools.readthedocs.io/en/latest/content/installation.html
-  * padillalab (tpadillabena)
+  * padillalab
 <code>
@@ Line 33: / Line 44: @@
 # ls /share/apps/CENTOS6/python/2.7.9/bin/macs2
 /share/apps/CENTOS6/python/2.7.9/bin/macs2
+# or change python version to
+which python: /share/apps/CENTOS7/python/3.8.3/bin/python
+pip list: Package         Version
+--------------- -------
+bowtie          0.11.0
+MACS2           2.2.7.1
+numpy           1.18.5
+picard          2.3.2
+scipy           1.4.1
@@ Line 191: / Line 212: @@
 ====== Python3 ======
-Added to both versions\\
+  * Python 3.8.3
+  * /share/apps/CENTOS7/python/3.8.3  (so queues mwgpu, exx96)
+  * most packages found on this page
+  * picked up mpirun from miniconda2 apps (that may be a problem)
+<code>
+# needed for _ctypes to compile properly for https retrievals
+ yum install openssl openssl-devel
+ yum install yum install libffi-devel libffi
+  tar -xvf ../Python-3.8.3.tgz
+  cd Python-3.8.3/
+  ./configure --prefix=/share/apps/CENTOS7/python/3.8.3 --enable-optimizations
+  make
+  make install
+  pip3 install --upgrade pip
+# make symbolic links python3->pyhton, pip3->pip
+# drop all the packages in from previous versions
+https://dokuwiki.wesleyan.edu/doku.php?id=cluster:73#python3
+</code>
+Added to both 3.5.2 versions\\
 pycuda\\
  --- //[[hmeij@wesleyan.edu|Henk]] 2020/02/18 08:45//
@@ Line 909: / Line 954: @@
 ====== Gromacs =====
+Gromacs-2020.3 for GPU (RTX2080S)
+  * ''exx96'' queue
+  * see /share/apps/CENTOS7/gromacs/2020.3.install for env
+  * change env in script below
+Gromacs-2018 for GPU (RTX2080S)
+  * ''exx96'' queue
+  * **/zfshomes/hmeij/k20redo/run.rtx sample script**
@@ Line 1016: / Line 1070: @@
 [[https://dokuwiki.wesleyan.edu/doku.php?id=cluster:119|GPU version]]
-  * program: mcc, math, mathematica (v 11.0.1)
+  * program: mcc, math, mathematica (v 12.1)
-  * install dir: /share/apps/CENTOS6/mathematica/11.0.1
+  * install dir: /share/apps/CENTOS6/mathematica/12.1
   * docs: on the web
   * www:  http://www.wolfram.com/mathematica/
+  * Campus wide license, use any queue.
   * Parallel kernels
@@ Line 1026: / Line 1081: @@
     * http://reference.wolfram.com/mathematica/guide/ParallelComputing.html
-Job submissions should go to queue //mathematica// which containes licensed nodes (all the hp12 nodes for now).
 ====== Eigen =====
@@ Line 1210: / Line 1265: @@
 ====== LAMMPS CPU/GPU ======
+** CPU only OR CPU+GPU**  for ''exx96'' queue - CentOS7
+  * git download of latest did not work (lots of fortran errors?)
+  * latest tarball is still 3Mar2020 (?)
+  * added user-eaxc (but had to remove package python)
+  * please read **/share/apps/CENTOS7/lammps/3Mar2020.install **
+<code>
+# CPU
+/share/apps/CENTOS7/lammps/3Mar2020/lmp_serial+user-omp+user-reaxc
+/share/apps/CENTOS7/lammps/3Mar2020/lmp_mpi+user-omp+user-reaxc
+# CPU+GPU
+/share/apps/CENTOS7/lammps/3Mar2020/lmp_mpi-single-single_kokkos_cuda_mpi+user-omp+user-reaxc
+# env
+ 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 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 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:$LD_LIBRARY_PATH
+</code>
+** CPU ONLY v2 CentOS7 **
+  * add user-opm on next compile ... done ... binaries with string +user-omp
+  * /share/apps/CENTOS7/lammps/3Mar2020
+    * lmp_serial, lmp_mpi
+  * //latest// openmpi version as of today v4.0.4
+  * consult this file for packages installed and compilations
+    * /share/apps/CENTOS7/lammps/03Mar2020.install
+    * should work only on exx96 and mwgpu queues
+<code>
+ export PATH=/share/apps/CENTOS7/openmpi/4.0.4/bin:$PATH
+ export LD_LIBRARY_PATH=/share/apps/CENTOS7/openmpi/4.0.4/lib:$LD_LIBRARY_PATH
+</code>
 ** CPU ONLY **
@@ Line 1222: / Line 1322: @@
 ** CPU or GPU **
+  * added lmp_mpi-single-single+user-omp
   * /share/apps/CENTOS7/lammps/3Mar2020
@@ Line 1228: / Line 1330: @@
     * /share/apps/CENTOS7/lammps/03Mar2020.install
     * CUDA_ARCH = sm_75 (probably only works on queue exx96)
-  * sample script in ~hmeij/k20redo/run.rtx.lammps
+  * **sample script in ~hmeij/k20redo/run.rtx.lammps**
 <code>

DokuWiki

User Tools

Site Tools

Differences

Page Tools