updated pages

softwares/abaqus.md

@@ -128,15 +128,15 @@ across three nodes i.e. 72 cores. Little is gained beyond three nodes.
 ## Job Submission Example
 
 Abaqus is installed under **/ichec/packages/abaqus** and is accessible
-via environment modules with the most recent being the default version:
+via environment modules. On **Kay** the most recent being the default version is:
 
 `module load abaqus/6.14-6`
 
 or
 
 `module load abaqus/2018`
 
-One Meluxina we have 2023 installed.
+On **Meluxina** we have 2023 installed.
 
 `module load abaqus/2023`
 

softwares/abinit.md

@@ -25,8 +25,8 @@ N/A.
 
 ### For Kay
 
-```bash -l
-#!/bin/bash
+```bash
+#!/bin/bash -l
 # All the information about queues can be obtained using 'sinfo'
 # PARTITION AVAIL  TIMELIMIT  
 # DevQ         up    1:00:00   
@@ -61,8 +61,8 @@ mpirun -np 80 abinit < MyInputFile > MyOutputFile
 
 ### For Meluxina
 
-```bash -l
-#!/bin/bash
+```bash
+#!/bin/bash -l
 # Slurm flags
 #SBATCH -p cpu
 #SBATCH -N 1

softwares/amber.md

@@ -5,25 +5,26 @@
 
 `Kay: 16 / 18`
 
-`LXP: `
+`LXP: 22`
 
 ```
 
 ## Description
 
-Amber (Assisted Model Building with Energy Refinement), is a general purpose molecular mechanics/dynamics suite which uses analytic potential
+AMBER (Assisted Model Building with Energy Refinement)[^1], is a general purpose molecular mechanics/dynamics suite which uses analytic potential
 energy functions, derived from experimental and ab initio data, to refine macromolecular conformations.
 
 ## License
 
 ICHEC has acquired a site license for the Amber 16 (AmberTools 17) and Amber 18 (AmberTools 18) packages.
 
-## Thin Component
+## Job Submission example
 
 A job submission example is as follows.
 
+### On Kay
 ```bash
-#!/bin/sh
+#!/bin/bash -l
 # All the informations about queues can be obtained by 'sinfo'
 # PARTITION AVAIL  TIMELIMIT 
 # DevQ         up    1:00:00
@@ -58,6 +59,40 @@ module load gcc/8.2.0
 srun $AMBERHOME/bin/pmemd.MPI -O -i mdin.in -o mdout.out -p prmtop.prm -c inpcrd.rst -ref inpcrd.rst -x trjfile.trj -inf file.info -r file.rst7
 ```
 
+
+### On Meluxina
+
+```bash
+#!/bin/bash -l
+
+# Slurm flags
+#SBATCH -p cpu
+#SBATCH --qos default
+#SBATCH -N 2
+#SBATCH --job-name=jobname 
+#SBATCH --hint=nomultithread
+#SBATCH -t 60:00:00
+
+# Charge job to myaccount
+#SBATCH -A your_project
+
+# Write stdout+stderr to file
+#SBATCH -o output.txt
+
+# Mail me on job start & end
+#SBATCH --mail-user=your_email
+#SBATCH --mail-type=BEGIN,END
+
+cd $SLURM_SUBMIT_DIR
+
+module load amber/22-foss-2023a-CUDA-12.2
+
+srun $AMBERHOME/bin/pmemd.MPI -O -i mdin.in -o mdout.out -p prmtop.prm -c inpcrd.rst -ref inpcrd.rst -x trjfile.trj -inf file.info -r file.rst7
+```
+
+
+
+
 ## Benchmarks
 
 Version: Amber 18
@@ -77,6 +112,10 @@ Merit: ns/day (the higher the better)
 
 Clearly, the speedup is enormous on GPU compared to CPUs.
 
+
+```{figure} https://www.ichec.ie/sites/default/files/inline-images/Cellulose_NPTPerformance_0.png
+:align: center
+```
 ```{mermaid}
 xychart-beta
     title "Performace vs Resource"
@@ -85,13 +124,12 @@ xychart-beta
     bar [0.15, 0.28, 4.27, 6.99, 84.71, 85.94]
 ```
 
-```{figure} https://www.ichec.ie/sites/default/files/inline-images/Cellulose_NPTPerformance_0.png
-:align: center
-```
-
 
 ### Additional Notes
 
 To use Amber 18 load the relevant environment module: module load
 amber/18. Use pmemd.MPI for MPI-only version, pmemd.cuda for CUDA-only
 version, pmemd.cuda.MPI for CUDA+MPI version.
+
+
+[^1]: See the AMBER wiki page [here](https://en.wikipedia.org/wiki/AMBER).

softwares/gaussian.md

@@ -5,7 +5,7 @@
 
 `Kay: 09e01 / 16b01`
 
-`LXP: `
+`LXP: 16c02`
 
 ```
 
@@ -43,10 +43,13 @@ case. Of course you can fine tune these values.
 
 `%mem=100000mb`
 
+## Job submission example
 An example of submission slurm script:
 
+### On Kay
+
 ```bash
-#!/bin/sh
+#!/bin/bash -l
 # All the information about queues can be obtained using 'sinfo'
 # PARTITION AVAIL  TIMELIMIT  
 # DevQ         up    1:00:00   
@@ -80,16 +83,51 @@ module load gaussian/16b01
 g16 < input.gjf > output.log   
 ```
 
+### On Meluxina
+
+```bash
+#!/bin/bash -l
+# Slurm flags
+#SBATCH -p cpu
+#SBATCH --qos default
+#SBATCH --hint=nomultithread
+#SBATCH -N 1
+#SBATCH -t 12:00:00
+#SBATCH --job-name=myjobname
+
+# Charge job to myproject 
+#SBATCH -A myproject
+
+# Write stdout+stderr to file
+#SBATCH -o output.txt
+
+# Mail me on job start & end
+#SBATCH --mail-user=myemailaddress@universityname.ie
+#SBATCH --mail-type=BEGIN,END
+
+cd $SLURM_SUBMIT_DIR
+echo $GAUSS_SCRDIR
+
+module load gaussian/16c02
+
+g16 < input.gjf > output.log
+```
+
+
 ## Additional notes
 
 The Gaussian module sets the **GAUSS_SCRDIR** to the correct location
 for that node type. Jobs can use very large RWF files as the scratch
 space is provided by a large high-performance shared volume including
 the follow in your submission script:
 
+
+For Meluxina, use RAMDISK as there are no local disks on compute nodes:
+
 ```bash
-export GAUSS_SCRDIR=/scratch/local/
+export GAUSS_SCRDIR="/dev/shm"
 ```
+For Kay, you can give path to any folder you have access to.
 
 Further information can be obtained at [www.gaussian.com](http://www.gaussian.com/ "Gaussian homepage").
 

softwares/gcc.md

@@ -5,7 +5,7 @@
 
 `Kay: 8.2.0, 9.4.0`
 
-`LXP: `
+`LXP: 12.3`
 
 ```
 
@@ -49,8 +49,12 @@ We save this in the C++ file named `helloworld.cpp`.  The program is compiled a
 This outputs an executable file called helloworld.  Here we have shown how to compile on the login node (i.e. at the command prompt when logged in to Kay), but be aware for very intensive compilations you should compile in a script and submit a job for this.  However, for most
 situations it is fine to compile on the login node.  Next, we show a script that can be used to run the program on Kay, on a single node.
 
+## Job submission example
+
+### On Kay
+
 ```bash
-#!/bin/sh
+#!/bin/bash -l
 # All the information about queues can be obtained using 'sinfo'
 # PARTITION AVAIL  TIMELIMIT 
 # DevQ         up    1:00:00  
@@ -80,6 +84,35 @@ cd $SLURM_SUBMIT_DIR
 
 module load gcc/9.4.0
 
+./helloworld
+```
+
+### On Meluxina
+
+```bash
+#!/bin/bash -l
+# Slurm flags
+#SBATCH -p cpu
+#SBATCH --qos default
+#SBATCH --hint=nomultithread
+#SBATCH -N 1
+#SBATCH -t 00:05:00
+#SBATCH --job-name=gccrun
+
+# Charge job to myproject
+#SBATCH -A MyProject
+
+# Write stdout+stderr to file
+#SBATCH -o output.txt
+
+# Mail me on job start & end
+#SBATCH --mail-user=myEmail@domain.ie
+#SBATCH --mail-type=BEGIN,END
+
+cd $SLURM_SUBMIT_DIR
+
+module load gcc/12.3.0
+
 ./helloworld
 ```