From: barate Date: Thu, 5 Dec 2013 12:55:57 +0000 (+0000) Subject: Update doc X-Git-Tag: V7_3_0b1 X-Git-Url: http://git.salome-platform.org/gitweb/?a=commitdiff_plain;h=6a0f897c7eac72aadffbd5eb98fb3f084c814564;p=modules%2Fjobmanager.git Update doc --- diff --git a/doc/images/jobmanager_job_workflow_3.png b/doc/images/jobmanager_job_workflow_3.png index 7b25398..48508f0 100644 Binary files a/doc/images/jobmanager_job_workflow_3.png and b/doc/images/jobmanager_job_workflow_3.png differ diff --git a/doc/images/jobmanager_job_workflow_4.png b/doc/images/jobmanager_job_workflow_4.png index 6e2e147..49f7ca7 100644 Binary files a/doc/images/jobmanager_job_workflow_4.png and b/doc/images/jobmanager_job_workflow_4.png differ diff --git a/doc/job.rst b/doc/job.rst index ecb98a8..b5639e1 100644 --- a/doc/job.rst +++ b/doc/job.rst @@ -61,10 +61,13 @@ The second table below describes the attributes of computation requirements. duration is set to 0, the time will be set to the default value of the batch queue selected. **Number of cpu** It's the number of cpus/cores requested. -**Memory** It's the amount of memory per node expected. +**Memory** It's the amount of required memory. It is generally specified **per node**. + With some batch mangers, it is possible to specify the required memory + **per core** (only available with SLURM for now). **Queue** It's optional. It permits to choose a specific batch queue on the targeted cluster. If it is not defined, most of the batch systems will affect your job to the queue that fits with the other attributes requirements. +**Exclusive** It indicates if the job can share nodes with other jobs or not. ======================== ============================================================================== Job's states diff --git a/doc/jobmanager_gui.rst b/doc/jobmanager_gui.rst index 7269491..dca1fe1 100644 --- a/doc/jobmanager_gui.rst +++ b/doc/jobmanager_gui.rst @@ -88,14 +88,12 @@ will be executed. You can also define the batch queue that you want to use. **Create wizard page 3** -The fourth page (see :ref:`figure_jobmanager_job_workflow_3`) permits to define the batch parameters related -to the job. It's mandatory to define the *Remote work directory* that is the directory where the job will be executed. -Note that if a work directory is defined for the selected resource, the work directory for the job will be -created as a subdirectory of the resource work directory by default. The name of this subdirectory will -contain the current date and time. Of course, you can override this default value. -Input files defined in page 4 will be copied in this work directory. You also have to define the maximum duration +The fourth page (see :ref:`figure_jobmanager_job_workflow_3`) allows to define the resource requirements +and limits for the job. You can define the maximum duration (set 0 hours and 0 minutes to use the default value for the queue), -the amount of memory needed and the number of processors/cores of the job. +the amount of memory needed (per node or per core if the batch manager supports it), the number of +processors/cores for the job, and whether it should use nodes exclusively (not sharing them with other +jobs). .. _figure_jobmanager_job_workflow_3: @@ -106,8 +104,14 @@ the amount of memory needed and the number of processors/cores of the job. The fifth page (see :ref:`figure_jobmanager_job_workflow_4`) permits to add the input and output files. Input files are files located into the user -computer that have to be transferred into the execution resource. In this page, you could also define the result -directory where job's results and logs will be copied. +computer that have to be transferred into the execution resource. Output files will be copied to the user +computer at the end of the job. +It's mandatory to define the *Remote working directory* that is the directory where the job will be executed. +Note that if a working directory is defined for the selected resource, the working directory for the job will be +created as a subdirectory of the resource working directory by default. The name of this subdirectory will +contain the current date and time. Of course, you can override this default value. +Input files will be copied in this working directory. +In this page, you can also define the result directory where the job's results and logs will be copied. .. _figure_jobmanager_job_workflow_4: