Tutorial: Using WestGrid. Drew Leske Compute Canada/WestGrid Site Lead University of Victoria



Similar documents
Using WestGrid. Patrick Mann, Manager, Technical Operations Jan.15, 2014

Introduction to Running Hadoop on the High Performance Clusters at the Center for Computational Research

Getting Started with HPC

Work Environment. David Tur HPC Expert. HPC Users Training September, 18th 2015

Miami University RedHawk Cluster Working with batch jobs on the Cluster

Linux für bwgrid. Sabine Richling, Heinz Kredel. Universitätsrechenzentrum Heidelberg Rechenzentrum Universität Mannheim. 27.

Beyond Windows: Using the Linux Servers and the Grid

Introduction to Sun Grid Engine (SGE)

An Introduction to High Performance Computing in the Department

New High-performance computing cluster: PAULI. Sascha Frick Institute for Physical Chemistry

Introduction to Linux and Cluster Basics for the CCR General Computing Cluster

Agenda. Using HPC Wales 2

Using Parallel Computing to Run Multiple Jobs

1.0. User Manual For HPC Cluster at GIKI. Volume. Ghulam Ishaq Khan Institute of Engineering Sciences & Technology

Grid Engine Users Guide p1 Edition

HPC at IU Overview. Abhinav Thota Research Technologies Indiana University

High Performance Computing

Grid Engine Basics. Table of Contents. Grid Engine Basics Version 1. (Formerly: Sun Grid Engine)

User s Manual

HPC system startup manual (version 1.30)

High Performance Computing with Sun Grid Engine on the HPSCC cluster. Fernando J. Pineda

Hodor and Bran - Job Scheduling and PBS Scripts

SGE Roll: Users Guide. Version Edition

Linux command line. An introduction to the Linux command line for genomics. Susan Fairley

Ra - Batch Scripts. Timothy H. Kaiser, Ph.D. tkaiser@mines.edu

Using the Yale HPC Clusters

Batch Scripts for RA & Mio

Introduction to the SGE/OGS batch-queuing system

Tutorial Guide to the IS Unix Service

Job scheduler details

NEC HPC-Linux-Cluster

Streamline Computing Linux Cluster User Training. ( Nottingham University)

Job Scheduling with Moab Cluster Suite

How To Run A Steady Case On A Creeper

Quick Tutorial for Portable Batch System (PBS)

Grid 101. Grid 101. Josh Hegie.

The Asterope compute cluster

Running applications on the Cray XC30 4/12/2015

HPCC USER S GUIDE. Version 1.2 July IITS (Research Support) Singapore Management University. IITS, Singapore Management University Page 1 of 35

Cluster Computing With R

Manual for using Super Computing Resources

Linux Overview. Local facilities. Linux commands. The vi (gvim) editor

PBS Tutorial. Fangrui Ma Universit of Nebraska-Lincoln. October 26th, 2007

Command Line Crash Course For Unix

Job Scheduling Explained More than you ever want to know about how jobs get scheduled on WestGrid systems...

Parallel Debugging with DDT

SLURM: Resource Management and Job Scheduling Software. Advanced Computing Center for Research and Education

High-Performance Reservoir Risk Assessment (Jacta Cluster)

SLURM: Resource Management and Job Scheduling Software. Advanced Computing Center for Research and Education

University of Toronto

Training Day : Linux

Secure Shell. The Protocol

CPSC 226 Lab Nine Fall 2015

How To Run A Tompouce Cluster On An Ipra (Inria) (Sun) 2 (Sun Geserade) (Sun-Ge) 2/5.2 (

Introduction to SDSC systems and data analytics software packages "

An Introduction to Using the Command Line Interface (CLI) to Work with Files and Directories

Linux Labs: mini survival guide

Introduction to Supercomputing with Janus

Birmingham Environment for Academic Research. Introduction to Linux Quick Reference Guide. Research Computing Team V1.0

sftp - secure file transfer program - how to transfer files to and from nrs-labs

The CNMS Computer Cluster

Data management on HPC platforms

Lab 1 Beginning C Program

WinSCP PuTTY as an alternative to F-Secure July 11, 2006

File transfer clients manual File Delivery Services

Using the Yale HPC Clusters

Martinos Center Compute Clusters

Using NeSI HPC Resources. NeSI Computational Science Team

The Maui High Performance Computing Center Department of Defense Supercomputing Resource Center (MHPCC DSRC) Hadoop Implementation on Riptide - -

Running on Blue Gene/Q at Argonne Leadership Computing Facility (ALCF)

Guide to the Configuration and Use of SFTP Clients for Uploading Digital Treatment Planning Data to ITC

Caltech Center for Advanced Computing Research System Guide: MRI2 Cluster (zwicky) January 2014

Installing and running COMSOL on a Linux cluster

Introduction to MSI* for PubH 8403

NYUAD HPC Center Running Jobs

Introduction to parallel computing and UPPMAX

The Moab Scheduler. Dan Mazur, McGill HPC Aug 23, 2013

Programming for GCSE Topic H: Operating Systems

File Transfer Examples. Running commands on other computers and transferring files between computers

Tutorial-4a: Parallel (multi-cpu) Computing

Introduction to Running Computations on the High Performance Clusters at the Center for Computational Research

Overview. Remote access and file transfer. SSH clients by platform. Logging in remotely

Unix Sampler. PEOPLE whoami id who

High Performance Computing Facility Specifications, Policies and Usage. Supercomputer Project. Bibliotheca Alexandrina

1 Basic commands. 2 Terminology. CS61B, Fall 2009 Simple UNIX Commands P. N. Hilfinger

Thirty Useful Unix Commands

SLURM Workload Manager

The RWTH Compute Cluster Environment

ELEC 377. Operating Systems. Week 1 Class 3

Introduction to Unix Tutorial

OLCF Best Practices (and More) Bill Renaud OLCF User Assistance Group

UMass High Performance Computing Center

Transcription:

Tutorial: Using WestGrid Drew Leske Compute Canada/WestGrid Site Lead University of Victoria

Fall 2013 Seminar Series Date Speaker Topic 23 September Lindsay Sill Introduction to WestGrid 9 October Drew Leske Tutorial: Using WestGrid 23 October Jonatan Aronsson Tutorial: Introduction to the WestGrid Development Environment 6 November Fiona Brinkman Case Study: Genomics, Bioinformatics and HPC How Computational Analyses Are Transforming Infectious Disease Control More information on these and other seminars offered: https://www.westgrid.ca/support/training

WestGrid User Basics To use WestGrid systems effectively, you will need to know: * Where to get help and information * Which systems are suited to your project * How to log on to those systems * Basic Linux commands * How to define and submit batch jobs

Finding Information and Getting Help * The WestGrid website: www.westgrid.ca * Guidance on choosing systems and running jobs, to information about specific systems * System health and upcoming maintenance events * WestGrid Support: support@westgrid.ca * For everything from account problems to parallelization questions and code optimization advice * No question too big or too small These are the most important items to take with you today.

Choosing a System The WestGrid website describes each computing facility and its size, architecture, memory, interconnect, and associated storage: https://www.westgrid.ca/support/systems Some systems will be better suited to your project than others. As well, some software is only available on certain systems. We can help you find the best system for your needs.

Choosing a System: Here Are a Few System Cores Memory Interconnect Storage Hermes 2112 24 GB/node 2 x GigE, 10:1 IB 1.2 PB Nestor 2304 24 GB/node IB QDR 1.2 PB Hungabee 2048 16 TB shared IB QDR 405 TB Silo n/a n/a n/a 3.15 PB These four systems represent, in general terms, a general purpose system appropriate for serial jobs; a cluster with a high- speed interconnect, suitable for parallel jobs; a shared memory system for problems requiring large amounts of memory; and a storage site.

Connecting to WestGrid Your workstation Scheduler Login nodes Nestor

Connecting: Software You Will Need * Access to WestGrid systems is via Secure Shell (ssh) * Linux and Mac clients are included in OS * Windows: PuTTy, WinSSH * File transfer via Secure Copy (scp) or Secure FTP (sftp) * Linux and Mac clients included in OS * Windows: WinSCP, FileZilla * Grid tools also available Everything you need to know: https://www.westgrid.ca/support/quickstart/new_users

Basic Linux Commands * List directory contents: ls, ls -a, ls l * Create, remove directory: mkdir <dir>, rmdir <dir> * Change to directory, change to parent: cd <dir>, cd.. * Copy files: cp <src> <tgt> * Copy directory: cp r <src> <tgt> * Move file or directory: mv <file> <dir> * Rename file or directory: mv <oldfile> <newfile> * Remove file: rm <file> * View file: less <file>, cat <file> Everything you need to know: https://www.westgrid.ca/support/quickstart/new_users

Job Basics * Login nodes are for data management, editing and compiling source code, quick tests, and job management * The real work is done on worker nodes * Requests are submitted to the batch system and enter into an appropriate queue * Jobs are dispatched to worker nodes by the scheduler according to their priority mainly FairShare

Job Basics: Job Dispatch Your workstation Scheduler Login nodes Nestor

Job Basics: FairShare * Everybody gets a fair share based on allocation (if any) and usage * In essence, a job whose owner has had little usage over the past while will have higher priority than a job whose owner has been a heavy user hence, fair share * Some groups are given a bigger share RAC allocations

Job Basics: Job Definition * A batch job is defined by a script with special directives embedded that tell the cluster what s required for the job * Memory * Cores * Wall time * If your job exceeds these resources, your job may be terminated before completion :- (

Job Basics: Essential Commands * Submit a job: qsub <script> * Check status of jobs: qstat <job> * Check scheduling: showq * Delete a job: qdel <job> * When will my job start? showstart <job> * How do I use that command? man qstat * How do I use showq? showq --help Everything you need to know: https://www.westgrid.ca/support/quickstart/new_users

Job Basics: hello.pbs #!/bin/bash #PBS -l procs=1 #PBS -j oe #PBS -W Output_Path=$HOME/20131009/${PBS_JOBID%%.*} date echo "Hello, world!" echo "Am having a wonderful time in $(/bin/hostname)." echo "Love, $(whoami)"

Job Basics: Submitting hello.pbs * Here s where we submit the job to the cluster: westgrid# qsub hello.pbs 16363886.moab01.westgrid.uvic.ca * Here we can check the status of the job: westgrid# qstat 16363886 Job id Name User Time Use S Queue -------------------- ---------------- --------------- -------- - ----- 16363886.moab01 hello.pbs dleske 0 Q hermes * If you get unknown job ID, job has completed

Job Basics: Results! * When the job has completed, the output files you specified in the job script will contain the results * For example: Wed Oct 9 08:48:51 PDT 2013 Hello, world! Am having a wonderful time in hermes0195. Love, dleske * Whoop! Science! * There may be other output in these files provided by the batch system

Job Basics: Your First Jobs * Everything you need to get started is at: * https://www.westgrid.ca/support/quickstart/new_users * Run a couple of goofy little test jobs to get familiar with how the system works * qsub, qstat, showq, qdel * Something didn t work? * Job output usually provides the best clues * E- mail support@westgrid.ca BUT PLEASE

Job Basics: HELP! * If your job failed and you can t figure out what went wrong, send a note to support@westgrid.ca * Please include essential details: * The name of the system you are using * The job ID * Your WestGrid user ID * Also anything else we may need to know to solve your real problem.

Recap: WestGrid User Basics To use WestGrid systems effectively, you will need to know: * Where to get help and information * Which systems are suited to your project * How to log on to those systems * Basic Linux commands * How to define and submit batch jobs

Information and Help The most important things to take away with you today: * WestGrid website: www.westgrid.ca * WestGrid Support: support@westgrid.ca You may also have local support at your institution. Don t be shy. We are here to support and enable you and your work.

Thanks for coming!