To enable and sustain large clusters, you need to tune
LSF for efficient querying, dispatching, and event log management.
Manage scheduling performance LSB_MAX_JOB_DISPATCH_PER_SESSION in lsf.conf and MAX_SBD_CONNS in lsb.params are set automatically during mbatchd startup to enable the fastest possible job dispatch.
Manage the number of pending reasons
Condense all the host-based pending reasons
into one generic pending reason for efficient, scalable management
of pending reasons.