a job's completion before allocating already released resources to pending
    jobs. This can be used to reduce fragmentation of resources.

  documentation, read config, show config, and run PrologSlurmctld when a job
  is allocated resources (startup synchronization still needed).

    https://computing.llnl.gov/linux/slurm/power_save.html or 
    "man slurm.conf" (SuspendProgram and related parameters) for more 
    information.
This is the final installment of the work: update some documenation, 
    increase default ResumeRate, and reduce frequency of retrying 
    batch launch state check.

    "man slurm.conf" (SuspendProgram and related parameters) for more 
    information.
NOTE: the step create logic needs to be modify to return EAGAIN or the 
    like until the node's state NOT_RESPONDING flag gets cleared.

    longer needed.

See NEWS for identification of all changes

  respect to SlurmdTimeout parameter.

added some documentation for not running slurmdbd for jobcomp plugin and redundancy for the accounting and jobcomp plugins

    separate flags for job, partition, and node data. See "man slurm.conf"
    for details.

Chris Morrone.

    (--task-mem will continue to be accepted for now, but is not documented).
    Replace DefMemPerTask and MaxMemPerTask with DefMemPerCPU, DefMemPerNode,
    MaxMemPerCPU and MaxMemPerNode in slurm.conf. Allocate a job's memory memory
    at the same time that processors are allocated based upon the --mem or
    --mem-per-cpu option rather than when job steps are initiated.

    getting information for all jobs. Improved performance of some commands.
    NOTE: Change in RPC means all nodes in the cluster should be updated
    at the same time.

    domain names (by Vijay Ramasubramanian, University of Maryland).

    partition's size and/or time limits rather than leaving them queued for a
    later change in the partition's limits. NOTE: Not reported by
    "scontrol show config" to avoid changing RPCs. It will be reported in 
    SLURM version 1.4.