From 096892f1150b58ec61ded1b596addff6db8b163f Mon Sep 17 00:00:00 2001 From: gwenwhite <66272666+gwenwhite@users.noreply.github.com> Date: Fri, 25 Jun 2021 13:53:43 -0600 Subject: [PATCH 1/2] Update Running_jobs_on_our_clusters.md --- wiki/pages/Running_jobs_on_our_clusters.md | 61 ++++++++++++++++++++++ 1 file changed, 61 insertions(+) diff --git a/wiki/pages/Running_jobs_on_our_clusters.md b/wiki/pages/Running_jobs_on_our_clusters.md index aeb5a70..3e3531c 100644 --- a/wiki/pages/Running_jobs_on_our_clusters.md +++ b/wiki/pages/Running_jobs_on_our_clusters.md @@ -10,6 +10,67 @@ If you are unsure what 'TPS' stands for and why we want so many of them, you sho Also, you'll need accounts on our supercomputing clusters - take a look at [Accounts for computer clusters](Accounts_for_computer_clusters.md) to get set up on Fry and Kestrel. +# Order of operations +### 1. Pull in the planckton image +In the home directory of your cluster, create an ```images``` directory +``` +cd $HOME +mkdir images +``` +Inside the images directory, pull in the [planckton](https://github.com/cmelab/planckton) image +``` +cd images +singularity pull docker://cmelab/planckton_gpu_v0.4.0.sif +export PLANCKTON_SIMG=$(pwd)/planckton_gpu_v0.4.0.sif +``` + + +### 2. Clone signac-flow +Inside of your project workspace clone [signac-flow](https://github.com/glotzerlab/signac-flow.git) +``` +cd $PROJECT +git clone https://github.com/glotzerlab/signac-flow.git +``` +Inside of the signac-flow clone directory, _____ +``` +cd signac-flow +git fetch origin pull/441/head:pr441 +git checkout pr441 +pip install . +``` + + +### 3. Clone planckton-flow +In the home of your project workspace, create a directory called ```jobs``` and clone [planckton-flow](https://github.com/cmelab/planckton-flow) into it. +``` +cd $HOME +mkdir jobs +cd jobs +git clone https://github.com/cmelab/planckton-flow.git +``` +Go into the planckton-flow clone directory, and set up the planckton-flow environment inside of it +``` +cd planckton-flow +conda env create -f environment.yml +conda activate planckton-flow +``` + +### 4. Run a job +Edit the init.py file in the src directory to the variables you want to see ran. Use python to initialize the jobs and then submit the project.py file to run the jobs using the variables you set. Check the status of you job by running the squeue command. +``` +vim src/init.py +python src/init.py +python src/project.py submit +squeue -u +``` + + +### Celebrate the fact that you just ran a job!!! +``` +echo "WOOHOO" +``` + + # What is a job scheduler and which one am I using? # From c3fdc56fd9a6c7a4703f6732d16f9e505ecd22ce Mon Sep 17 00:00:00 2001 From: gwenwhite <66272666+gwenwhite@users.noreply.github.com> Date: Tue, 29 Jun 2021 14:04:04 -0600 Subject: [PATCH 2/2] Update Running_jobs_on_our_clusters.md Added a description to the signac flow section. --- wiki/pages/Running_jobs_on_our_clusters.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/wiki/pages/Running_jobs_on_our_clusters.md b/wiki/pages/Running_jobs_on_our_clusters.md index 3e3531c..e6bf51a 100644 --- a/wiki/pages/Running_jobs_on_our_clusters.md +++ b/wiki/pages/Running_jobs_on_our_clusters.md @@ -31,7 +31,7 @@ Inside of your project workspace clone [signac-flow](https://github.com/glotzerl cd $PROJECT git clone https://github.com/glotzerlab/signac-flow.git ``` -Inside of the signac-flow clone directory, _____ +Inside of the signac-flow clone directory, pull commit #441 then checkout the corresponding branch. ``` cd signac-flow git fetch origin pull/441/head:pr441