HP XC System 4.x Software guide

74

Image 74

HP XC System 4.x Software manual

Contents

HP XC System Software XC User Guide Page Table of Contents Configuring Your Environment with Modulefiles Compiling and Linking Serial Applications Debugging Applications Srun Roles Srun Modes 107 115 125 131 11-1 List of Figures10-1 List of Tables List of Examples New and Changed Information in This Edition About This DocumentIntended Audience Typographic Conventions HP XC and Related HP Products Information Related Information HP XC System Software Users Guide Https//computing.llnl.gov/linux/slurm/documentation.html Http//systemimager.org Manpages Http//www-unix.mcs.anl.gov/mpi $ man -k keyword HP Encourages Your Comments$ man discover $ man 8 discover Page Page Determining the Node Platform Only to the administrator of the HP XC systemNode Specialization /proc/cpuinfo file is dynamic SAN Storage Storage and I/OFile System Local Storage HP XC System Interconnects System Interconnect NetworkFile System Layout Network Address Translation NAT Determining System Configuration InformationUser Environment 1 LVS Parallel Applications CommandsApplication Development Environment Slurm Run-Time EnvironmentSerial Applications Load Sharing Facility LSF HP-MPI Requested by the HP-MPI mpirun commandHow LSF and Slurm Interact Nodes for the job Components, Tools, Compilers, Libraries, and Debuggers Page Using the Secure Shell to Log Using the SystemLVS Login Routing Logging In to the System Getting Information About Resources IntroductionGetting Information About Queues $ bqueues Performing Other Common User Tasks $ lsid Getting System Help and InformationDetermining the LSF Cluster Name and the LSF Execution Host $ man sinfo Configuring Your Environment with Modulefiles Overview of Modules Supplied Modulefiles Supplied Modulefiles HP-MPI Viewing Loaded Modulefiles Modulefiles Automatically Loaded on the SystemViewing Available Modulefiles Loading a Modulefile Modulefile Conflicts Unloading a ModulefileViewing Modulefile-Specific Help Creating a Modulefile $ module load modules $ man modulefile Developing Applications Application Development Environment Overview MPI Compiler Compiler CommandsCompilers Interrupting a Job Setting Debugging OptionsExamining Nodes and Partitions Before Running Jobs Developing Serial Applications Building Serial Applications Developing Parallel ApplicationsSerial Application Build Environment Parallel Application Build Environment OpenMP ModulefilesHP-MPI Pthreads PGI Fortran and C/C++ Compilers MPI LibraryIntel Fortran and C/C++Compilers GNU C and C++ Compilers Building Parallel Applications Designing Libraries for the CP4000 Platform To compile and link a C application using the mpicc commandDeveloping Libraries To build a 64-bit application, you might enter Linkcommand 32-bit -L/opt/mypackage/lib/i686 -lmystuffLinkcommand 64-bit -L/opt/mypackage/lib/x8664 -lmystuff Developing Applications Submitting Jobs Submitting a Serial Job Using LSFSubmitting a Serial Job with the LSF bsub Command Overview of Job Submission Example 5-1 Submitting a Job from the Standard Input Example 5-2 Submitting a Serial Job Using LSFSubmitting a Serial Job Through Slurm Only $ bsub -I srun hostname Submitting a Non-MPI Parallel Job Following is the command line used to compile this programSubmitting a Parallel Job To submit a parallel job Example 5-5 Submitting a Non-MPI Parallel Job $ bsub -n4 -I srun hostname Example 5-7 Submitting an MPI Job $ bsub -n4 -I mpirun -srun ./helloworld Arguments for the Slurm External Scheduler Nodelist=list of nodes $ bsub -n 10 -ext SLURMnodelist=n1-10 srun hostname $ bsub -n 10 -ext SLURMconstraint=dualcore -I srun hostname Submitting a Batch Job or Job ScriptExample 5-14 Submitting a Job Script $ bqueues -l dualcore grep Slurm $ bsub -n4 -ext SLURMnodes=4 -I ./myscript.sh $ bsub -n4 -I ./myscript.sh Use the squeue command to acquire information on the jobs Using a Script to Submit Multiple JobsUsing a Makefile to Submit Multiple Jobs Submitting Multiple MPI Jobs Across the Same Set of Nodes $ tail 113.out Following command line makes the program and executes it$ cat mymake $ cat 117.out Type=SLINUX64 Submitting a Job from a Host Other Than an HP XC HostRunning Preexecution Programs $ bsub -R type=SLINUX64 -n4 -I srun hostnamePage Debugging Parallel Applications Debugging ApplicationsDebugging Serial Applications TotalView SSH and TotalView Setting Up TotalViewDebugging with TotalView Module load mpi module load totalview Setting TotalView Preferences Using TotalView with SlurmUsing TotalView with LSF $ srun -Nx-A $ mpirun -tv -srun application Debugging an Application $ mpirun -tv -srun -n2 ./Psimple Exiting TotalView $ scancel --user usernameDebugging Running Applications Run the applicationPage Monitoring Node Activity Xtools Utilities Running Performance Health Tests Where $ ovp --verify=perfhealth/cpuusage You can list the available tests with the ovp -lcommand$ ovp -l $ ovp --verbose --verify=perfhealth/cpuusage HOMEDIRECTORY/ovpn16mmddyy.log HOMEDIRECTORY/ovpn16mmddyyr1.log Page Building a Program Intel Trace Collector and HP-MPI Tuning ApplicationsUsing the Intel Trace Collector and Intel Trace Analyzer For more information, see the following Web site Running a Program Intel Trace Collector and HP-MPIExample 8-1 The vtjacobic Example Program Example 8-2 C Example Running the vtjacobic Example Program HP-MPI and the Intel Trace Collector Installation KitIntel Trace Collector and Analyzer with HP-MPI on HP XC Running a Program Following is a Fortran example called vtjacobif # bsub -n4 -I mpirun.mpich -np 2 ./vtjacobic Running a Program Across Nodes Using LSFVisualizing Data Intel Trace Analyzer and HP-MPI Page Introduction to Slurm Using SlurmLaunching Jobs with the srun Command Slurm Utilities Using the srun Command with LSF Monitoring Jobs with the squeue CommandUsing the srun Command with HP-MPI Srun Roles and Modes Example 9-5cancels all pending jobs Terminating Jobs with the scancel CommandGetting System Information with the sinfo Command Security Fault ToleranceJob Accounting $ sinfo -R Using LSF Information for LSF Overview of LSF Integrated with Slurm $ squeue --jobs $SLURMJOBID Lsfadmin@n16 ~$ bsub -n4 -I srun hostname Example 10-1 Examples of LSF Job Launch Lshosts and lsload commands display for each of these items Differences Between LSF and LSF Integrated with Slurm Submitted. This job may run immediately, or it may run Job TerminologyConnection back to the terminal from which the job was $ ssh n15 lshosts Are submitted with the bsub -Icommand Batch jobs are submitted with the srun -bcommand. ByDefault, the output is written to Batch system scheduling policies Submitting Jobs Using LSF Integrated with Slurm in the HP XC EnvironmentUseful Commands Job Startup and Job Control LSF with Slurm Job Launch Exit Codes LSF-SLURM External SchedulerHow LSF and Slurm Launch and Manage a Job User logs in to login node n16 How LSF and Slurm Launch and Manage a Job Determining the LSF Execution Host Determining Available System Resources Getting Information About the LSF Execution Host Node Following example shows the output from the lshosts commandExamining System Core Status Examining System Queues Getting Information About JobsGetting Host Load Information Getting Information About the lsf Partition LSF job with Slurm allocated resources Getting Job Allocation InformationThis allocation string has the following values Than what the job requests Example 10-4 Job Allocation Information for a Finished Job Example 10-5 Using the bjobs Command Short OutputExamining the Status of a Job Output Provided by the bhist Command Example 10-6 Using the bjobs Command Long OutputExample 10-7 Using the bhist Command Short Output Viewing the Historical Information for a Job Translating Slurm and LSF JOBIDs Example 10-8 Using the bhist Command Long OutputUse the bjobs command to view the Slurm Jobid $ bsub -I -n4 -ext SLURMnodes=4 /bin/bash Working Interactively Within an Allocation$ sacct -j $ bjobs -l 124 grep slurm Alternatively, you can use the following Example 10-9 Launching an Interactive MPI JobThis example assumes 2 cores per node LSF Equivalents of Slurm srun Options LSF Equivalents of Slurm srun Options3describes the srun options and lists their LSF equivalents $ unset Slurmjobid $ unset Slurmnprocs Requests a specific list of nodes. Slurm Suppress informational message Running an X Terminal Session from a Remote Node Advanced TopicsEnabling Remote Execution with OpenSSH Determining IP Address of Your Local Machine Logging in to HP XC System Running an X terminal Session Using SlurmRunning an X terminal Session Using LSF Run the job on 1 core Using the GNU Parallel Make CapabilityOptions used in this command are Allocate 4 cores Monitors display server address $ cd subdir srun $MAKE $ cd subdir srun -n1 -N1 $MAKE -j4 Example Procedure $ make PREFIX=’srun -n1 -N1 MAKEJ=-j4 $ make PREFIX=srun -n1 -N1 MAKEJ=-j4 Local Disks on Compute NodesModified Makefile is invoked as follows Communication Between Nodes Using Mpich on the HP XC System11.5 I/O Performance Considerations Shared File View Bsub command launches the wrapper script Using Mpich with Slurm AllocationUsing Mpich with LSF Allocation Launching a Serial Interactive Shell Through LSF ExamplesBuilding and Running a Serial Application Examine the LSF execution host information Running LSF Jobs with a Slurm Allocation Request Launching a Parallel Interactive Shell Through LSF Example 2. Four Cores on Two Specific Nodes Examine the running jobs information $ hostname n16 $ srun hostname n5 $ bjobs Show the environment Submitting a Simple Job Script with LSFExamine the finished jobs information Display the script Submit the job Submitting an Interactive Job with LSFRun the job Show the job allocation Exit the pseudo-terminal Run some commands from the pseudo-terminalShow the Slurm job ID View the interactive jobs View the running job Submitting an HP-MPI Job with LSFView the node state $ bsub -n 8 -R ALPHA5 SLINUX64 \ -ext SLURMnodes=4-4 myjob Using a Resource Requirements String in an LSF CommandView the finished job 124 Glossary Network Availability set Fcfs Ipmi LVS PXE SVA Index Index LVS PGI 135