A number of properties that apply to the job as a whole
may be specified when the job is submitted.
Job Property
|
Description
|
Name
|
Text string that identifies/describes
the job.
|
Template
|
Template that controls job property
defaults and allowed values.
|
Type
|
Computing resource unit granularity:
Node, Socket, or Core.
|
Priority
|
Lowest, BelowNormal,
Normal, AboveNormal, and Highest.
When multiple jobs are ready to run, Job
Priority controls which jobs are run by the scheduler.
|
Units
|
The min/max number of computing
resource units that the job requires to run. The job will run on at
least the minimum (and no more than the maximum) number of resource
units. For example, in order to run four distributed engines, each requiring
four cores for multi-processing, set the number of job units to sixteen.
|
Exclusive
|
If true, no other jobs will be
scheduled on nodes allocated to this job.
The Exclusive property value allows the additional
control on the resources used by a job. For example, a job may be scheduled
which requires all memory on a host. When only some of the host cores
are available, the Exclusive property would prevent other jobs
from being scheduled on the unused cores.
|
Node Groups
|
Specifies the node groups on which
a job may be run.
Node groups can be created by the cluster administrator
in order to group together similar nodes. For example, a node group
“32GB Nodes” can be comprised of nodes that have 32GB of
available memory. As a result, jobs that are memory intensive can be
configured to run on the “32GB Nodes” group.
|
Node List
|
When specified, runs the job only
on nodes listed in the Node List. Additionally, other job properties
exist that can limit which nodes may be allocated to a job:
|
Minimum Processors
|
Minimum number of cores per node.
|
Minimum Memory
|
Minimum memory per node, expressed
in megabytes (MB).
|
Node Limits
|
Specifies the min/max node limits
on which a job may be run.
— Minimum node limits ensures that a job will
have sufficient resources (nodes and memory) in order to efficiently
run.
— Maximum node limits optimize the use of
cluster resources.
|