OSCER users,
OSCER TRAINING WORKSHOPS IN 2024 -- live via Zoom!
(1) Introduction to Linux/Unix Commands and
Running Batch Jobs on OSCER's Supercomputer
[Once in each of Jan, Feb, March, Apr, July, Sep, Oct, Nov.]
(2) Running Python Jobs and Using Jupyter for Small Jobs
on OSCER's Supercomputer
[Once in each of Jan, March, Sep and Nov.]
Details are below. Offerings are subject to change without
notice.
The OSCER Team ([log in to unmask])
---
(1) Introduction to Linux/Unix Commands and
Running Batch Jobs on OSCER's Supercomputer
Dates
Tue Jan 16 9:00-11:00am
Thu Feb 15 1:00-3:00pm
Thu March 14 1:00-3:00pm
Wed Apr 10 9:00-11:00am
Wed July 31 9:00-11:00am
Thu Sep 12 1:00-3:00pm
Wed Oct 16 9:00-11:00am
Thu Nov 14 1:00-3:00pm
Topics
* Logging in to OSCER's supercomputer
* Linux/Unix Basics
-- Basic Linux/Unix commands: pwd, ls, mkdir, cd, cp, cat,
chmod, man, exit, etc
-- Basic text editing for creating/modifying batch script
files, small input text files, etc
* OSCER supercomputer storage
-- Navigating your /home and /scratch directories
-- File uploading/downloading to/from supercomputer storage
-- OURdisk basics
* Batch Jobs
-- Overview of batch computing
-- Creating a Slurm batch script
-- Basic Slurm batch commands for submitting a batch job,
watching its status, viewing/following its output/error files,
cancelling it
-- Modules: searching, loading, listing, unloading/purging
PLUS, extras if time permits
* Useful Linux/Unix Command Line Skills
-- Copy-and-paste in the Linux/Unix terminal window
-- Re-running/modifying previous commands using the arrow keys
-- Filename completion using the tab key
-- Wildcards
* Batch Job Skills
-- Modifying a pending batch job (Slurm scontrol command)
-- Viewing info about past batch jobs (Slurm sacct command)
---
(2) Running Python Jobs and Using Jupyter for Small Jobs
on OSCER's Supercomputer
Dates
Wed Jan 31 9:00-11:00am
Tue March 26 9:00-11:00am
Tue Sep 24 9:00-11:00am
Thu Nov 28 1:00-3:00pm
Topics
* Creating a virtual Python environment in your home directory
and installing your preferred Python packages
* Small interactive Python runs using Jupyter
(JupyterLab/Jupyter Notebooks)
* Python for Machine Learning (especially on GPUs)
-- Module(s) for GPU-enabled Python ML (cuDNN/CUDA/Python)
-- Creating, submitting and monitoring a Python GPU ML
batch script
PLUS, extra if time permits
* Tensorflow links
|
