Hp XC System 2.x Software Instrukcja Użytkownika Strona 72

  • Pobierz
  • Dodaj do moich podręczników
  • Drukuj
  • Strona
    / 154
  • Spis treści
  • BOOKMARKI
  • Oceniono. / 5. Na podstawie oceny klientów
Przeglądanie stron 71
Table 6-1: SLURM Commands (cont.)
Command
Function
sinfo
Reports the state of partitions and nodes managed by SLURM. It has a wide variety
of filtering, sorting, and formatting options. sinfo displays a summary of available
partition and node (not job) information (such as partition names, nodes/partition,
and CPUs/node).
scontrol
Is an administrative tool used to view or modify the SLU RM state. Typically, users
do not need to access this command. Therefore, the scontrol command can only
be executed as user root. Refer to the HP XC System Software Administration
Guide for information about using this command.
The -help command option also provides a brief summary of SLURM options. Note that
command options are not case sensitive.
6.3 Accessing the SLURM Manpages
You
can also view online descriptions of these com m ands by accessing the SLURM m anpages.
Man
pages are provided for all SLURM commands and API functions. If SLURM manpages
are
not already available in your MANPATH environment variable, you can set and export
the
masfollows:
$ MANPATH=$MANPATH:/opt/hptc/man
$ export MANPATH
Yo u can now access the SLURM manpages with the standard man command. For example:
$ ma
n srun
6.4
Launching Jobs with the srun Command
The srun command submits jo bs to run u nder SLURM management. Jobs can be submitted t o
run in parallel on mu ltiple compute nodes. srun is used to submit a job for execution, allocate
resources, a tt ach to an existing allocation , or i nit iate job steps. srun can perform the following:
•Sub
mit a batch job and then terminate
Submit an interactive job and then persist to sh epherd the job as it runs
Allocate resources to a shell and then spaw n that shell for use in running subordinate jo bs
Jobs can be submitted for immediate execution or later execution (batch). srun has a wide
variety of options to specify resource requirements, including: minimum and maximum node
count, processor count, specific nodes to use o r not use, and specific node characteristics (so
much memory, disk space, certain required features). Besides securing a resource allocation,
srun is used to initiate job steps. These job steps can execute sequentially or in parallel on
independent or shared nodes w ithin the job’s node allocation.
Ex
ample 6-1: Simple Launch of a Serial Program
$ sr
un -n2 -l hostname
0: n
1
1: n
1
6.4.1 The srun Roles and Modes
The srun comman d submits jobs to ru n under SLURM management. T h e srun command can
perform many roles in launching and m anaging your job. srun also provides several disti nct
usage modes to accommodate the roles it performs.
6-2 Using SLURM
Przeglądanie stron 71
1 2 ... 67 68 69 70 71 72 73 74 75 76 77 ... 153 154

Komentarze do niniejszej Instrukcji

Brak uwag