README 3.69 KB
Newer Older
Autosubmit is a tool to create, manage and monitor experiments by using 
configured Computing Clusters, HPC's and Supercomputers remotely via ssh.
Muhammad Asif's avatar
 
Muhammad Asif committed

HOW TO DEPLOY/SETUP AUTOSUBMIT FRAMEWORK
========================================
- Autosubmit has been tested:
  with the following Operating Systems:
   * Linux Debian
  on the following HPC's/Clusters:
   * Ithaca (IC3 machine)
   * MareNostrum (BSC machine)
   * MareNostrum3 (BSC machine)
   * HECToR (EPCC machine)
   * Lindgren (PDC machine)
   * C2A (ECMWF machine)
   * ARCHER (EPCC machine)
- Pre-requisties: These packages (bash, python2, python-argparse, python-dateutil, python-pyparsing, python-numpy,
python-pydot, python-matplotlib, sqlite3, git-scm > 1.8.2) must be available at local machine.
And the machine is also able to access HPC's/Clusters via password-less ssh.
- Install Autosubmit
   > pip install autosubmit
  or download, unpack and "python setup.py install"

- Create a repository for experiments: Say for example "/cfu/autosubmit" then
  edit the repository path (LOCAL_ROOT_DIR) into autosubmit/config/dir_config.py
- Create a blank database: Say for example "autosubmit.db" at above created repository:
   > cp autosubmit/database/data/autosubmit.sql /cfu/autosubmit/
   > cd /cfu/autosubmit
   > sqlite3 autosubmit.db
   sqlite3>.read autosubmit.sql
   > chmod 775 autosubmit.db
  then edit the database file path and name (DB_DIR, DB_FILE, DB_NAME) into autosubmit/config/dir_config.py
HOW TO USE AUTOSUBMIT
=====================
To run AUTOSUBMiT experiments at CFU a production environment is set up at the local virtual machine "enterprise".
> python expid.py --new --HPC ithaca --description "experiment is about..."
Muhammad Asif's avatar
 
Muhammad Asif committed

Muhammad Asif's avatar
Muhammad Asif committed
Say for example, "cxxx" is 4 character based expid generated by system automatically.
First character "c" represents the platform such as "i" for ithaca, "b" for bsc, 
"h" for hector, "l" for lindgren, "e" for ecmwf and "m" for marenostrum3 etc. While rest
of three characters "xxx" are to represent unique alphanumeric identity for the experiment.
Muhammad Asif's avatar
 
Muhammad Asif committed

Muhammad Asif's avatar
Muhammad Asif committed
> vi /cfu/autosubmit/cxxx/conf/expdef_cxxx.conf
Muhammad Asif's avatar
Muhammad Asif committed
> vi /cfu/autosubmit/cxxx/conf/autosubmit_cxxx.conf
Muhammad Asif's avatar
 
Muhammad Asif committed

Muhammad Asif's avatar
Muhammad Asif committed
> python create_exp.py cxxx
Muhammad Asif's avatar
 
Muhammad Asif committed

> ssh enterprise
Muhammad Asif's avatar
Muhammad Asif committed
> nohup python autosubmit.py cxxx >& cxxx_01.log &
Cautions: 
- Before launching autosubmit check the following stuff:
> ssh ithaca # say for example similarly check other HPC's where password-less ssh is feasible
- After launching autosubmit, one must be aware of login expeiry limit and policy (if applicable for any HPC) 
and renew the login access accordingly (by using token/key etc) before expiry.
HOW TO MONITOR EXPERIMENT
=========================
Muhammad Asif's avatar
 
Muhammad Asif committed

Muhammad Asif's avatar
Muhammad Asif committed
> python monitor.py -e cxxx -j job_list -o pdf
Muhammad Asif's avatar
Muhammad Asif committed
> python monitor.py -e cxxx -j job_list -o png
Above generated plot with date & time stamp can be found at:
Muhammad Asif's avatar
Muhammad Asif committed
/cfu/autosubmit/cxxx/plot/cxxx_date_time.pdf
Muhammad Asif's avatar
Muhammad Asif committed
/cfu/autosubmit/cxxx/plot/cxxx_date_time.png
HOW TO RESTART EXPERIMENT
=========================
Muhammad Asif's avatar
Muhammad Asif committed
> python recovery.py -e cxxx -j job_list -g # getting/fetching completed files
Muhammad Asif's avatar
Muhammad Asif committed
> python recovery.py -e cxxx -j job_list -s # saving the pickle file
Muhammad Asif's avatar
Muhammad Asif committed
> nohup python autosubmit.py cxxx >& cxxx_02.log &


HOW TO RERUN/EXTEND EXPERIMENT
==============================

> ssh enterprise
Muhammad Asif's avatar
Muhammad Asif committed
> vi /cfu/autosubmit/cxxx/conf/expdef_cxxx.conf # modify RERUN, CHUNKLIST
Muhammad Asif's avatar
Muhammad Asif committed
> python create_exp.py cxxx
Muhammad Asif's avatar
Muhammad Asif committed
> nohup python autosubmit.py cxxx >& cxxx_03.log &
Muhammad Asif's avatar
Muhammad Asif committed
> python monitor.py -e cxxx -j rerun_job_list -o pdf
Muhammad Asif's avatar
Muhammad Asif committed
> python recovery.py -e cxxx -j rerun_job_list -g 
Muhammad Asif's avatar
Muhammad Asif committed
> python recovery.py -e cxxx -j rerun_job_list -s