HP XC System 3.x Software - Manual (Page 83)

10 Using LSF-HPC
The Load Sharing Facility (LSF-HPC) from Platform Computing Corporation is a batch system resource
manager used on the HP XC system.
On an HP XC system, a job is submitted to LSF-HPC, which places the job in a queue and allows it to run
when the necessary resources become available. In addition to launching jobs, LSF-HPC provides extensive
job management and information capabilities. LSF-HPC schedules, launches, controls, and tracks jobs that
are submitted to it according to the policies established by the HP XC site administrator.
Two types of LSF are available for installation on the HP XC:
• LSF-HPC
This product is the popular batch system produced by Platform Computing Corporation that has
become an industry standard. For full information about LSF-HPC, see the standard LSF-HPC
documentation set, which is described in the
“Related Software Products and Additional Publications”
section of this manual. LSF-HPC manpages are also available online on the HP XC system.
• LSF-HPC integrated with SLURM (LSF-HPC)
This product is the High Performance LSF product from Platform Computing Corporation that has
been integrated with SLURM to take advantage of SLURM's scalable, efficient resource management
and parallel job support.
The intended primary use of the HP XC system determined which of these LSF-HPC products was installed.
This chapter introduces you to LSF-HPC integrated with SLURM in the HP XC environment. It provides
an overview of how LSF-HPC integrated with SLURM works, and discusses some of the features and
differences of LSF-HPC compared to LSF-HPC integrated with SLURM on an HP XC system.
This chapter also contains an important discussion of how LSF-HPC and SLURM work together to provide
the HP XC job management environment. A description of SLURM is provided in
Chapter 9 (page 79) .
This chapter addresses the following topics:
• “Information for LSF-HPC” (page 83)
• “Overview of LSF-HPC Integrated with SLURM” (page 84)
• “Differences Between LSF-HPC and LSF-HPC Integrated with SLURM” (page 85)
• “Job Terminology” (page 86)
• “Using LSF-HPC Integrated with SLURM in the HP XC Environment” (page 87)
• “Submitting Jobs” (page 88)
• “How LSF-HPC and SLURM Launch and Manage a Job” (page 88)
• “Determining the LSF Execution Host” (page 90)
• “ Determining Available System Resources” (page 90)
• “Getting Information About Jobs” (page 92)
• “Translating SLURM and LSF-HPC JOBIDs” (page 96)
• “Working Interactively Within an Allocation” (page 97)
• “LSF-HPC Equivalents of SLURM srun Options” (page 99)
10.1 Information for LSF-HPC
The information for LSF-HPC is provided in the LSF documentation. This documentation is on the HP
XC installation disk and manpages are online.
LSF-HPC is installed and configured on all nodes of the HP XC system by default. Nodes without the
compute
role are closed with '0' job slots available for use.
The LSF environment is set up automatically for the user on login; LSF commands and their manpages
are readily accessible:
• The bhosts command is useful for viewing LSF batch host information.
• The lshosts command provides static resource information.
• The lsload command provides dynamic resource information.
10.1 Information for LSF-HPC 83