User Tools

Site Tools


hpc_concepts

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
hpc_concepts [2009/11/17 11:19] 172.26.0.166hpc_concepts [2010/05/22 14:19] (current) – external edit 127.0.0.1
Line 1: Line 1:
-**Message Passing Interface (MPI): The Concept**+==== Message Passing Interface (MPI): The Concept ====
  
 ---- ----
Line 9: Line 9:
 If you are simply looking for how to run an MPI application, you probably want to use a command line of the following form: If you are simply looking for how to run an MPI application, you probably want to use a command line of the following form:
  
-           shell$ mpirun [ -np X ] [ --hostfile <filename> ]  <program>+<file>        shell$ mpirun [ -np X ] [ --hostfile <filename> ]  <program>
  
        This will run X copies of <program> in your current run-time environment (if running under a        This will run X copies of <program> in your current run-time environment (if running under a
Line 16: Line 16:
        the use of a hostfile, or will default to running all X copies on the localhost), scheduling        the use of a hostfile, or will default to running all X copies on the localhost), scheduling
        (by  default)  in  a  round-robin  fashion  by CPU slot.  See the rest of this page for more        (by  default)  in  a  round-robin  fashion  by CPU slot.  See the rest of this page for more
-       details. +       details. </file> 
-===== Installation =====+=== Installation === 
 +---- 
 + 
 <file>$ wget http://www.open-mpi.org/software/ompi/v1.3/downloads/openmpi-1.3.3  <file>$ wget http://www.open-mpi.org/software/ompi/v1.3/downloads/openmpi-1.3.3 
 $ tar xfz openmpi-1.3.3.tar.gz  $ tar xfz openmpi-1.3.3.tar.gz 
Line 34: Line 37:
 http://www.cs.wisc.edu/condor/downloads-v2/download.pl http://www.cs.wisc.edu/condor/downloads-v2/download.pl
  
-**Sun Grid Engine (SGE)**+==== Sun Grid Engine (SGE) ====
 ---- ----
  
Line 42: Line 45:
  
  
-**SLURM: A Highly Scalable Resource Manager**+==== SLURM: A Highly Scalable Resource Manager ====
  
 SLURM is an open-source resource manager designed for Linux clusters of all sizes. It provides three key functions. First it allocates exclusive and/or non-exclusive access to resources (computer nodes) to users for some duration of time so they can perform work. Second, it provides a framework for starting, executing, and monitoring work (typically a parallel job) on a set of allocated nodes. Finally, it arbitrates contention for resources by managing a queue of pending work. https://computing.llnl.gov/linux/slurm/ SLURM is an open-source resource manager designed for Linux clusters of all sizes. It provides three key functions. First it allocates exclusive and/or non-exclusive access to resources (computer nodes) to users for some duration of time so they can perform work. Second, it provides a framework for starting, executing, and monitoring work (typically a parallel job) on a set of allocated nodes. Finally, it arbitrates contention for resources by managing a queue of pending work. https://computing.llnl.gov/linux/slurm/
  
  
-**TORQUE Resource Manager**+==== TORQUE Resource Manager ====
  
 TORQUE is an open source resource manager providing control over batch jobs and distributed compute nodes. It is a community effort based on the original *PBS project and, with more than 1,200 patches, has incorporated significant advances in the areas of scalability, fault tolerance, and feature extensions contributed by NCSA, OSC, USC , the U.S http://www.clusterresources.com/pages/products/torque-resource-manager.php TORQUE is an open source resource manager providing control over batch jobs and distributed compute nodes. It is a community effort based on the original *PBS project and, with more than 1,200 patches, has incorporated significant advances in the areas of scalability, fault tolerance, and feature extensions contributed by NCSA, OSC, USC , the U.S http://www.clusterresources.com/pages/products/torque-resource-manager.php
  
-**Platfrom LSF**+==== Platfrom LSF ====
  
 [[platform_lsf|LSF]] is implemented as a resource manager for the HPC together with SGE.  [[platform_lsf|LSF]] is implemented as a resource manager for the HPC together with SGE. 
hpc_concepts.1258456743.txt.gz · Last modified: 2010/05/22 14:19 (external edit)