Hewlett-Packard Company Palo Alto, California
June Product Version
Page
Contents
5.4
Developing Applications
Debugging Applications
Tuning Applications
Using Slurm
Using HP-MPI
Using LSF
6.4
Using HP Mlib
Glossary Index Examples
Advanced Topics
Examples
Tables
Figures
Page
Document Organization
About This Document
Intended Audience
HP XC Information
Linux Administration Handbook
QuickSpecs for HP XC System Software
HP XC Program Development Environment
Supplementary Information
For More Information
HP Message Passing Interface
HP Mathematical Library
Http//supermon.sourceforge.net
Manpages
Http//systemimager.org
Http//sourceforge.net/projects/modules
Http//www-unix.mcs.anl.gov/mpi
Related Information
Http//linuxvirtualserver.org
Bold text
# cd /opt/hptc/config/sbin
Typographical Conventions
Ctrl/x
HP Encourages Your Comments
Page
System Architecture
Overview of the User Environment
Operating System
Node Specialization
File System
Storage and I/O
SAN Storage
Local Storage
File System Layout
System Interconnect Network
User Environment
Commands
Network Address Translation NAT
1 LVS
Serial Applications
Application Development Environment
Parallel Applications
Slurm
Run-Time Environment
Load Sharing Facility LSF-HPC
How LSF-HPC and Slurm Interact
HP-MPI
Components, Tools, Compilers, Libraries, and Debuggers
8Overview of the User Environment
Configuring Your Environment with Modulefiles
Using the System
LVS Login Routing
Using ssh to Log
2Using the System
Modulefile Sets the HP XC User Environment
Supplied Modulefiles
Supplied Modulefiles
Viewing Available Modulefiles
Modulefiles Automatically Loaded on the System
Viewing Loaded Modulefiles
Loading a Modulefile
Unloading a Modulefile
Automatically Loading a Modulefile at Login
Modulefile Conflicts
Loading a Modulefile for the Current Session
Viewing Modulefile-Specific Help
Creating a Modulefile
$ module load modules $ man modulefile
$ module help totalview
Introduction
Launching and Managing Jobs Quick Start
Getting Information About Queues
Getting Information About Resources
Launching Jobs
Getting Information About the System’s Partitions
Submitting a Serial Job
Example 2-1 Submitting a Serial Job
Submitting a Non-MPI Parallel Job
Using Slurm Options with the LSF External Scheduler
Example 2-2 Submitting a Non-MPI Parallel Job
$ bsub -n4 -I srun hostname
$ bsub -n4 Mpirun -srun ./helloworld
Submitting an MPI Job
Example 2-4 Running an MPI Job with LSF
$ bsub -I -n4 myjobscript.sh
Submitting a Batch Job or Job Script
Example 2-6 Submitting a Job Script
Getting System Help and Information
Performing Other Common User Tasks
$ man -k keyword
$ man sinfo
Page
Overview
Developing Applications
Standard Linux Compilers
Using Compilers
Intel Compilers
PGI Compilers
Checking Nodes and Partitions Before Running Jobs
Setting Debugging Options
Interrupting a Job
Developing Serial Applications
Developing Parallel Applications
Using Mlib in Serial Applications
Serial Application Build Environment
Building Serial Applications
Modulefiles
Parallel Application Build Environment
HP-MPI
OpenMP
Mlib Math Library
Quadrics Shmem
MPI Library
$ mpicc object1.o ... -pthread -o myapp.exe
PGI Fortran and C/C++ Compilers
Intel Fortran and C/C++Compilers
GNU C and C++ Compilers
GNU Parallel Make
Building Parallel Applications
Reserved Symbols and Names
Compiling and Linking Non-MPI Applications
Compiling and Linking HP-MPI Applications
$ mpicc -c -g foo.c
Developing Libraries
Designing Libraries for XC4000
Advanced Topics
Using the GNU Parallel Make Capability
Example 3-1 Directory Structure
Example 3-2 Recommended Directory Structure
$ cd subdir srun -n1 -N1 $MAKE -j4
$ cd subdir srun $MAKE
Example Procedure
$ make PREFIX=’srun -n1 -N1’ MAKEJ=’-j4’
$ make PREFIX=’srun -n1 -N1 MAKEJ=’-j4’
3 I/O Performance Considerations
Local Disks on Compute Nodes
Shared File View
Private File View
Communication Between Nodes
Page
Debugging Serial Applications
Debugging Applications
Debugging Parallel Applications
TotalView
SSH and TotalView
Setting Up TotalView
Debugging with TotalView
Using TotalView with LSF-HPC
Using TotalView with Slurm
$ srun -Nx-A $ mpirun -tv -srun application
$ bsub -nx-ext SLURMnodes=x \ -Is /usr/bin/xterm
$ totalview
Starting TotalView for the First Time
TotalView Preferences Window
Preferences window, click on the Launch Strings tab
Debugging Applications
$ mpirun -tv -srun -n2 ./Psimple
Debugging an Application
$ mpicc -g -o Psimple simple.c -lm
TotalView Process Window Example
$ mpirun -srun -n2 Psimple
Debugging Running Applications
$ squeue
$ scancel --user username
Exiting TotalView
Page
Using the Intel Trace Collector/Analyzer
Tuning Applications
Building a Program Intel Trace Collector and HP-MPI
Example
Example Running the vtjacobic Example Program
Running a Program Intel Trace Collector and HP-MPI
Visualizing Data Intel Trace Analyzer and HP-MPI
Slurm Commands
Using Slurm
Command Function
Introduction
Launching Jobs with the srun Command
Accessing the Slurm Manpages
Srun Roles and Modes
Example 6-1 Simple Launch of a Serial Program
Srun Modes
Srun Roles
Srun Run-Mode Options
Srun Signal Handling
Batch
Allocate
Srun Resource-Allocation Options
Part --partition=part
Cpt --cpus-per-task=cpt
Minutes --time=minutes
Nthreads --threads=nthreads
Srun Control Options
Srun I/O Options
Mode --input=mode
Mode --output=mode
Mode --error=mode
Label
Clist --constraint=clist
Srun Constraint Options
Mem=size
Contiguous=yesno
Mincpus=n
Vmem=size
Using srun with HP-MPI
Monitoring Jobs with the squeue Command
Using srun with LSF
Srun Environment Variables
Getting System Information with the sinfo Command
Killing Jobs with the scancel Command
Job Accounting
Fault Tolerance
Security
$ sinfo -R
Overview of LSF
Using LSF
Introduction to LSF in the HP XC Environment
Nodelist= list-of-nodes Exclude= list-of-nodes
Topology Support
$ bsub -n 10 -ext SLURMnodes=10exclude=n16 srun myapp
$ bsub -n 10 -ext SLURMnodes=10 srun myapp
$ bsub -Is hostname
How LSF and Slurm Launch and Manage a Job
$ bqueues -l normal grep Jobstarter
$ bsub -n4 -ext SLURMnodes=4 -o output.out ./myscript
How LSF-HPC and Slurm Launch and Manage a Job
Differences Between LSF on HP XC and Standard LSF
Determining Available System Resources
Determining Execution Host
Getting Status of LSF
Job Startup and Job Control
Getting Host Load Information
Getting Information About LSF-HPC Execution Host Node
Checking LSF System Queues
Submitting Jobs
Getting Information About the lsf Partition
$ sinfo -p lsf
Bsub bsuboptions jobname joboptions
Summary of the LSF bsub Command Format
Slurm Arguments Function
LSF-SLURM External Scheduler
Starting on lsfhost.localdomain n6
Submitting a Job in Parallel
Submitting a Serial Job
Submitting an HP-MPI Job
Example 7-5 Submitting an Interactive Serial Job
$ bsub -n4 -I mpirun -srun ./helloworld
Submitting a Batch Job or Job Script
Example 7-6 Submitting an HP-MPI Job
Example 7-8 Submitting a Batch Job Script
Examples
$ bsub -n4 -I ./myscript.sh
$ bsub -n4 -ext SLURMnodes=4 -I ./myscript.sh
Type=SLINUX64
Submitting a Job from a Non-HP XC Host
$ bsub -n4 -I ./myscript.sh -n8 -O
Getting Job Allocation Information
Getting Information About Jobs
Job Allocation Information for a Running Job
$ bsub -R type=SLINUX64 -n4 -I srun hostname
Job Allocation Information for a Finished Job
Example 7-13 Using the bjobs Command Short Output
Checking Status of a Job
Example 7-15 Using the bhist Command Short Output
Example 7-14 Using the bjobs Command Long Output
Output Provided by the bhist Command
Viewing a Job’s Historical Information
Submitting an Interactive Job to Launch the xterm Program
Example 7-16 Using the bhist Command Long Output
Working Interactively Within an LSF-HPC Allocation
Example 7-18 View Your Allocation in Slurm
Example 7-17 View Your Environment
Example 7-19 View Your Running Job in LSF
Example 7-20 View Job Details in LSF
Example 7-21 Running Jobs from an xterm Window
Submitting an Interactive Job to Launch a Shell
Example 7-22 Submitting an Interactive Shell Program
$ hostname $ srun hostname $ srun -n2 hostname
LSF Equivalents of Slurm srun Options
LSF Equivalents of Slurm srun Options
$ srun hostname n1 $ exit
Srun Option Description LSF Equivalent
24Using LSF
Control the parallel job
Page
Using HP-MPI
HP-MPI Directory Structure
Setting Environment Variables
Compiling and Running Applications
Building and Running an Example Application
Building and Running helloworld
Example Application helloworld
$ mpicc -o helloworld $MPIROOT/help/helloworld.c
$ $MPIROOT/bin/mpirun -srun -n4 helloworld
HP-MPI options allowed with -srun
Launching MPI Jobs
System Interconnect Selection
Creating Subshells and Launching Jobsteps
$ mpirun -srun -n6 -O -N2 -m cyclic ./a.out host1 rank1
$ mpirun -srun -n4 -N2 -O -m cyclic ./a.out host1 rank1
$ mpirun -subnet 192.168.1.1 -prot -srun -n4 ./a.out
Using LSF and HP-MPI
System Interconnect Support
$ /usr/sbin/ifconfig -a
MPI Versioning
Example 8-5 Allocating 12 Processors on 6 Nodes
Allowing Windows to Use Exclusive Locks
32-Bit Builds on XC4000
Truncated Messages
$ mpirun -TCP -srun -N8 ./a.out
Mpirun Command Options
$MPIROOT/bin/mpirun -v -prot -np 2 /path/to/program.x
Environment Variables
Export MPIPINPERCENTAGE=30
Export MPIPHYSICALMEMORY=1048576
Export MPIPAGEALIGNMEM=1
Export MPIMAXWINDOW=10
Export MPIUSELIBELAN=0
Mpich Object Compatibility
Export MPIUSELIBELANUSE=5
$MPIROOT/bin/mpirun.mpich -np 2 ./prog.x
At http//docs.hp.com
HP-MPI Documentation and Manpages
HP-MPI Manpage Categories
Category Manpages Description
Additional Information, Known Problems, and Work-arounds
Intel Compiler Notes
Using HP Mlib
Version 8 Fortran Compiler
Version 7 Fortran Compiler
Mlib and Module Files
HP Mlib for the HP XC6000 Platform
Platform Support
Library Support
Compiling and Linking
Using Intel Compilers with HP Mlib
Modulefiles and Mlib
Licensing
HP Mlib for the HP XC4000 Platform
Mlib Manpages
Linking SuperLUDIST
$ pgcc options file
6Using HP Mlib
Enabling Remote Execution with OpenSSH
Advanced Topics
Running an X Terminal Session from a Remote Node
Determining IP Address of Your Local Machine
Logging in to HP XC System
Running an X terminal Session Using Slurm
$ hostname
$ host mymachine
$ bjobs
Running an X terminal Session Using LSF
$ bsub -n4 -Ip srun -n1 xterm -display
Page
Launching a Serial Interactive Shell Through LSF
Examples
Building and Running a Serial Application
Example 1. Two Processors on Any Two Nodes
Running LSF Jobs with a Slurm Allocation Request
View the job
Launching a Parallel Interactive Shell Through LSF
Example 2. Four Processors on Two Specific Nodes
$ bsub -Is -n4 -ext SLURMnodes=4 /bin/bash
Check the running job’s information
Check the finished job’s information
Submitting a Simple Job Script with LSF
Show the environment
Display the script
Show the job allocation
Submitting an Interactive Job with LSF
Show the Slurm job ID
$ bsub -n8 -Ip /bin/sh
Submitting an HP-MPI Job with LSF
$ bjobs -l
View the running job
View the finished job
$ bsub -n 8 -R ALPHA5 SLINUX64 \ -ext SLURMnodes=4-4 myjob
Using a Resource Requirements String in an LSF Command
Page
Glossary
External network node
Extensible firmware interface
Fairshare
First come first served
Integrated Lights Out
Image server
Interconnect
Internet address
LSF master host
Network Information Services
Management Processor
Master host
Parallel application
Root Administration Switch
Resource manager role
Role
Glossary-6
Symmetric multiprocessing
Index-1
Index
Index-2
Gdb, 4-1 GNU
Index-3
Modulefile
Index-4
Resource manager, 7-1role Serial applications