You appear to have allocated all of the nodes memory to the first job. All jobs need to fit into memory then the jobs can be time sliced. I will try to clarify in the documentation.
Quoting Sefa Arslan <sefa.ars...@tubitak.gov.tr>: > > Yes I have: > > NodeName=mercan[5-192] Procs=24 Sockets=2 CoresPerSocket=12 > RealMemory=128000 State=UNKNOWN > PartitionName=test Nodes=mercan[7-192] Default=YES MaxTime=INFINITE > State=UP DefMemPerCPU=5300 Shared=FORCE:4 > > > scontrol show config: > Configuration data as of 2012-06-29T09:57:56 > AccountingStorageBackupHost = (null) > AccountingStorageEnforce = associations,limits > AccountingStorageHost = mercan5 > AccountingStorageLoc = N/A > AccountingStoragePort = 6819 > AccountingStorageType = accounting_storage/slurmdbd > AccountingStorageUser = N/A > AccountingStoreJobComment = YES > AuthType = auth/munge > BackupAddr = (null) > BackupController = (null) > BatchStartTimeout = 10 sec > BOOT_TIME = 2012-06-29T09:12:09 > CacheGroups = 1 > CheckpointType = checkpoint/none > ClusterName = linux > CompleteWait = 0 sec > ControlAddr = mercan5 > ControlMachine = mercan5 > CryptoType = crypto/munge > DebugFlags = (null) > DefMemPerCPU = 5300 > DisableRootJobs = NO > EnforcePartLimits = YES > Epilog = (null) > EpilogMsgTime = 2000 usec > EpilogSlurmctld = (null) > FastSchedule = 1 > FirstJobId = 100000 > GetEnvTimeout = 2 sec > GresTypes = (null) > GroupUpdateForce = 0 > GroupUpdateTime = 600 sec > HASH_VAL = Match > HealthCheckInterval = 0 sec > HealthCheckProgram = (null) > InactiveLimit = 0 sec > JobAcctGatherFrequency = 30 sec > JobAcctGatherType = jobacct_gather/linux > JobCheckpointDir = /var/slurm/checkpoint > JobCompHost = localhost > JobCompLoc = /var/log/slurm/job_completions > JobCompPort = 0 > JobCompType = jobcomp/filetxt > JobCompUser = root > JobCredentialPrivateKey = (null) > JobCredentialPublicCertificate = (null) > JobFileAppend = 0 > JobRequeue = 1 > JobSubmitPlugins = (null) > KillOnBadExit = 0 > KillWait = 30 sec > Licenses = (null) > MailProg = /bin/mail > MaxJobCount = 1000000 > MaxJobId = 4294901760 > MaxMemPerNode = UNLIMITED > MaxStepCount = 40000 > MaxTasksPerNode = 128 > MessageTimeout = 10 sec > MinJobAge = 300 sec > MpiDefault = none > MpiParams = (null) > NEXT_JOB_ID = 106036 > OverTimeLimit = 0 min > PluginDir = /usr/lib64/slurm > PlugStackConfig = /etc/slurm/plugstack.conf > PreemptMode = GANG,SUSPEND > PreemptType = preempt/partition_prio > PriorityDecayHalfLife = 00:00:00 > PriorityCalcPeriod = 00:05:00 > PriorityFavorSmall = 0 > PriorityMaxAge = 14-00:00:00 > PriorityUsageResetPeriod = NONE > PriorityType = priority/multifactor > PriorityWeightAge = 1000 > PriorityWeightFairShare = 10000 > PriorityWeightJobSize = 1000 > PriorityWeightPartition = 1000 > PriorityWeightQOS = 0 > PrivateData = none > ProctrackType = proctrack/cgroup > Prolog = (null) > PrologSlurmctld = (null) > PropagatePrioProcess = 0 > PropagateResourceLimits = (null) > PropagateResourceLimitsExcept = MEMLOCK > RebootProgram = (null) > ReconfigFlags = (null) > ResumeProgram = (null) > ResumeRate = 300 nodes/min > ResumeTimeout = 60 sec > ResvOverRun = 0 min > ReturnToService = 0 > SallocDefaultCommand = (null) > SchedulerParameters = (null) > SchedulerPort = 7321 > SchedulerRootFilter = 1 > SchedulerTimeSlice = 30 sec > SchedulerType = sched/builtin > SelectType = select/cons_res > SelectTypeParameters = CR_CPU_MEMORY > SlurmUser = root(0) > SlurmctldDebug = debug3 > SlurmctldLogFile = /var/log/slurm/slurmctld.log > SlurmSchedLogFile = (null) > SlurmctldPort = 6817 > SlurmctldTimeout = 300 sec > SlurmdDebug = debug3 > SlurmdLogFile = /var/log/slurm/slurmd.log > SlurmdPidFile = /var/run/slurmd.pid > SlurmdPort = 6818 > SlurmdSpoolDir = /tmp/slurmd > SlurmdTimeout = 300 sec > SlurmdUser = root(0) > SlurmSchedLogLevel = 0 > SlurmctldPidFile = /var/run/slurmctld.pid > SLURM_CONF = /etc/slurm/slurm.conf > SLURM_VERSION = 2.4.0-pre4 > SrunEpilog = (null) > SrunProlog = (null) > StateSaveLocation = /tmp > SuspendExcNodes = (null) > SuspendExcParts = (null) > SuspendProgram = (null) > SuspendRate = 60 nodes/min > SuspendTime = NONE > SuspendTimeout = 30 sec > SwitchType = switch/none > TaskEpilog = (null) > TaskPlugin = task/cgroup > TaskPluginParam = (null type) > TaskProlog = (null) > TmpFS = /tmp > TopologyPlugin = topology/none > TrackWCKey = 0 > TreeWidth = 50 > UsePam = 1 > UnkillableStepProgram = (null) > UnkillableStepTimeout = 60 sec > VSizeFactor = 0 percent > WaitTime = 0 sec > > > On 06/28/2012 05:53 PM, Moe Jette wrote: >> >> Do you have "Shared" option configured for the partitions as >> instructed here: >> http://www.schedmd.com/slurmdocs/gang_scheduling.html >> >> Quoting Sefa Arslan <sefa.ars...@tubitak.gov.tr>: >> >>> I have hollpwing configuration for Gang Scheduling: >>> >>> SchedulerTimeSlice = 30 sec >>> SchedulerType = sched/builtin >>> SelectType = select/cons_res >>> SelectTypeParameters = CR_CORE_MEMORY >>> PreemptMode = GANG,SUSPEND >>> PreemptType = preempt/partition_prio >>> >>> I have submitted to jobs to the same node with full number of cores. >>> The first job started to run immediately. The other is at pendig state. >>> >>> 105986 test sleep sefa PD 0:00 1 (Resources) >>> 105985 test sleep sefa R 5:05 1 mercan89 >>> >>> Then I have set the priority of second job to a more higher value ( >>> scontrol update JobID=105986 Priority=100000000 ). I was expecting >>> the second job to be started. job 105986 is still at pending state. >>> >>> How should I configure the slurm? >>> >>> One more question: Could any user (or the admin) migrate his/her own >>> running job from one node to another? How? >>> >>> >>> >>> >>> >
