LSF Version 7.3 - Administering Platform LSF

Controlling Processor Allocation Across Hosts
504 Administering Platform LSF
❖ Become unavailable to the current job
❖ Remain available to other jobs as either regular or first execution hosts
◆ You cannot specify first execution host candidates when you use the brun
command.
If the first execution host is incorrect at job submission, the job is rejected. If
incorrect configurations exist on the queue level, warning messages are logged and
displayed when LSF starts, restarts or is reconfigured.
Job chunking Specifying first execution host candidates affects job chunking. For example, the
following jobs have different job requirements, and is not placed in the same job
chunk:
bsub -n 2 -m "hostA! hostB hostC" myjob
bsub -n 2 -m "hostA hostB hostC" myjob
bsub -n 2 -m "hostA hostB! hostC" myjob
The requirements of each job in this example are:
◆ Job 1 must start on hostA
◆ Job 2 can start and run on hostA, hostB, or hostC
◆ Job 3 must start on hostB
For job chunking, all jobs must request the same hosts and the same first execution
hosts (if specified). Jobs that specify a host preference must all specify the same
preference.
Resource
reservation
If you specify first execution host candidates at the job or queue level, LSF tries to
reserve a job slot on the first execution host. If LSF cannot reserve a first execution
host job slot, it does not reserve slots on any other hosts.
Controlling Processor Allocation Across Hosts
Sometimes you need to control how the selected processors for a parallel job are
distributed across the hosts in the cluster.
You can control this at the job level or at the queue level. The queue specification is
ignored if your job specifies its own locality.
Specifying parallel job locality at the job level
By default, LSF does allocate the required processors for the job from the available
set of processors.
A parallel job may span multiple hosts, with a specifiable number of processes
allocated to each host. A job may be scheduled on to a single multiprocessor host
to take advantage of its efficient shared memory, or spread out on to multiple hosts
to take advantage of their aggregate memory and swap space. Flexible spanning
may also be used to achieve parallel I/O.
You are able to specify “select all the processors for this parallel batch job on the
same host”, or “do not choose more than
n processors on one host” by using the
span section in the resource requirement string (bsub -R or RES_REQ in the queue
definition in
lsb.queues).