SLURM is a resource manager for clusters of any size. SLURM maintains a queue of work (typically parallel jobs), allocates resources to those jobs, and will initiate and otherwise manage those jobs.What\'s New in This Release: [ read full changelog ]· Added support for smap to query off node name for display.· Slurmdbd modified to set user ID and group ID to SlurmUser if started as· user root.· Configuration parameter ControlMachine changed to accept multiple comma-· separated hostnames for support of some high-availability architectures.· ALTERED API CALL slurm_get_job_steps 0 has been changed to NO_VAL for both· job and step id to recieve all jobs/steps. Please make adjustments to· your code.· salloc\'s --wait= option deprecated by --immediate= option to· match the srun command.· Add new slurmctld list for node features with node bitmaps for simplified· scheduling logic.· Multiple features can be specified when creating a reservation. Use \"&\"· (AND) or \"|\" (OR) separators between the feature names.· Changed internal node name caching so that front-end mode would work with· multiple lines of node name definitions.· Add node state flag for power-up/configuring. Represented by \"#\" suffix· on the node state name (e.g. \"ALLOCATED#\") for command output.· Add CONFIGURING/CF job state flag for node power-up/configuring.· Modify job step cancel logic for scancel and srun (on reciept of SIGTERM· or three SIGINT) to immediately send SIGKILL to spawned tasks. Previous· logic would send SIGCONT, SIGTERM, wait KillWait seconds, SIGKILL.· Create a spank_get_item() option that the SPANK cpuset plugin can use to· discover the cpus that the select/cons_res plugin has allocated to a job.· Improve sview performance (outrageously) on very large machines.· Add support for licenses in resource reservation.· BLUEGENE - Jobs waiting for a block to boot will now be in Configuring· state.· bit_fmt now does not return brackets surrounding any set of data.