Knowledge Center Contents Previous Next |
lsb_parameterinfo()
Returns information about the LSF cluster.
DESCRIPTION
lsb_parameterinfo() gets information about the LSF cluster.
The static storage for the parameterInfo structure is reused on the next call.
SYNOPSIS
#include <lsf/lsbatch.h> struct parameterInfo *lsb_parameterinfo(char **names, int *numUsers, int options) struct parameterInfo { char *defaultQueues; char *defaultHostSpec; int mbatchdInterval; int sbatchdInterval; int jobAcceptInterval; int maxDispRetries; int maxSbdRetries; int preemptPeriod; int cleanPeriod; int maxNumJobs; float historyHours; int pgSuspendIt; char *defaultProject; int retryIntvl; int nqsQueuesFlags; int nqsRequestsFlags; int maxPreExecRetry; int eventWatchTime; float runTimeFactor; float waitTimeFactor; float runJobFactor; int eEventCheckIntvl; int rusageUpdateRate; int rusageUpdatePercent; int condCheckTime; int maxSbdConnections; int rschedInterval; int maxSchedStay; int freshPeriod; int preemptFor; int adminSuspend; int userReservation; float cpuTimeFactor; int fyStart; int maxJobArraySize; time_t exceptReplayPeriod; int jobTerminateInterval; int disableUAcctMap; int enforceFSProj; int enforceProjCheck; int jobRunTimes; int dbDefaultIntval; int dbHjobCountIntval; int dbQjobCountIntval; int dbUjobCountIntval; int dbJobResUsageIntval; int dbLoadIntval; int dbJobInfoIntval; int jobDepLastSub; int maxJobNameDep; char *dbSelectLoad; int jobSynJgrp; char *pjobSpoolDir; int maxUserPriority; int jobPriorityValue; int jobPriorityTime; int enableAutoAdjust; int autoAdjustAtNumPend; float autoAdjustAtPercent; int sharedResourceUpdFactor; int scheRawLoad; char *jobAttaDir; int maxJobMsgNum; int maxJobAttaSize; int mbdRefreshTime; int updJobRusageInterval; char *sysMapAcct; int preExecDelay; int updEventUpdateInterval; int resourceReservePerSlot; int maxJobId; char *preemptResourceList; int preemptionWaitTime; int maxAcctArchiveNum; int acctArchiveInDays; int acctArchiveInSize; float committedRunTimeFactor; int enableHistRunTime; #ifdef PS_SXNQS int nqsUpdateInterval; #endif int mcbOlmReclaimTimeDelay; int chunkJobDuration; int sessionInterval; int publishReasonJobNum; int publishReasonInterval; int publishReason4AllJobInterval; int mcUpdPendingReasonInterval; int mcUpdPendingReasonPkgSize; int noPreemptRunTime; int noPreemptFinishTime; char * acctArchiveAt; int absoluteRunLimit; int lsbExitRateDuration; int lsbTriggerDuration; int maxJobinfoQueryPeriod; int jobSubRetryInterval; int pendingJobThreshold; int maxConcurrentJobQuery; int minSwitchPeriod; int condensePendingReasons; int slotBasedParallelSched; int disableUserJobMovement; int detectIdleJobAfter; int useSymbolPriority; int JobPriorityRound; char* priorityMapping; int maxInfoDirs; int minMbdRefreshTime; int enableStopAskingLicenses2LS; int expiredTime; char* mbdQueryCPUs; char *defaultApp; int enableStream; char *streamFile; int streamSize; int syncUpHostStatusWithLIM; char *defaultSLA; int slaTimer; int mbdEgoTtl; int mbdEgoConnTimeout; int mbdEgoReadTimeout; int mbdUseEgoMXJ; int mbdEgoReclaimByQueue; int defaultSLAvelocity; char *exitRateTypes; float globalJobExitRate; int enableJobExitRatePerSlot; int enableMetric; int schMetricsSample; float maxApsValue; int newjobRefresh; int preemptJobType; char *defaultJgrp; int jobRunlimitRatio; int jobIncludePostproc; int jobPostprocTimeout; int sschedUpdateSummaryInterval; int sschedUpdateSummaryByTask; int sschedRequeueLimit; int sschedRetryLimit; int sschedMaxTasks; int sschedMaxRuntime; char *sschedAcctDir; int jgrpAutoDel; int maxJobPreempt; int maxJobRequeue; int noPreemptRunTimePercent; int maxStreamFileNum; int PrivilegedUserForceBkill; int intersectCandidateHosts; int enforceOneUGLimit; int logRuntimeESTExceeded; char* computeUnitTypes; float fairAdjustFactor; int noPreemptFinishTimePercent; int slotReserveQueueLimit; int maxJobPercentagePerSession; int useSuspSlots; };PARAMETERS
**names
Reserved but not used; supply NULL.
*numUsers
Reserved but not used; supply NULL.
options
Reserved but not used; supply 0.
parameterInfo structure
The parameterInfo structure contains the following fields:
defaultQueues
DEFAULT_QUEUE: A blank_separated list of queue names for automatic queue selection.
defaultHostSpec
DEFAULT_HOST_SPEC: The host name or host model name used as the system default for scaling CPULIMIT and RUNLIMIT.
mbatchdInterval
MBD_SLEEP_TIME: The interval in seconds at which the mbatchd dispatches jobs.
sbatchdInterval
SBD_SLEEP_TIME: The interval in seconds at which the sbatchd suspends or resumes jobs.
jobAcceptInterval
JOB_ACCEPT_INTERVAL: The interval at which a host accepts two successive jobs. (In units of SBD_SLEEP_TIME.)
maxDispRetries
MAX_RETRY: The maximum number of retries for dispatching a job.
maxSbdRetries
MAX_SBD_FAIL: The maximum number of retries for reaching an sbatchd.
preemptPeriod
PREEM_PERIOD: The interval in seconds for preempting jobs running on the same host.
cleanPeriod
CLEAN_PERIOD: The interval in seconds during which finished jobs are kept in core.
maxNumJobs
MAX_JOB_NUM: The maximum number of finished jobs that are logged in the current event file.
historyHours
HIST_HOURS: The number of hours of resource consumption history used for fair share scheduling and scheduling within a host partition.
pgSuspendIt
PG_SUSP_IT: The interval a host must be idle before resuming a job suspended for excessive paging.
defaultProject
The default project assigned to jobs.
retryIntvl
Job submission retry interval
nqsQueuesFlags
For Cray NQS compatiblilty only. Used by LSF to get the NQS queue information
maxPreExecRetry
The maximum number of times to attempt the preexecution command of a job from a remote cluster ( MultiCluster only)
eventWatchTime
Event watching Interval in seconds
runTimeFactor
Run time weighting factor for fairshare scheduling
waitTimeFactor
used for calcultion of the fairshare scheduling formula
runJobFactor
Job slots weighting factor for fairshare scheduling
eEventCheckIntvl
default check interval
rusageUpdateRate
sbatchd report every sbd_sleep_time
rusageUpdatePercent
sbatchd updates jobs jRusage in mbatchd if more than 10% changes
condCheckTime
time period to check for reconfig
maxSbdConnections
The maximum number of connections between master and slave batch daemons
rschedInterval
The interval for rescheduling jobs
maxSchedStay
Max time mbd stays in scheduling routine, after which take a breather
freshPeriod
During which load remains fresh
preemptFor
The preemption behavior, GROUP_MAX, GROUP_JLP, USER_JLP, HOST_JLU, MINI_JOB, LEAST_RUN_TIME
adminSuspend
Flags whether users can resume their jobs when suspended by the LSF administrator
userReservation
Flags to enable/disable normal user to create advance reservation
cpuTimeFactor
CPU time weighting factor for fairshare scheduling
fyStart
The starting month for a fiscal year
maxJobArraySize
The maximum number of jobs in a job array
exceptReplayPeriod
Replay period for exceptions, in seconds
jobTerminateInterval
The interval to terminate a job
disableUAcctMap
User level account mapping for remote jobs is disabled
enforceFSProj
If set to TRUE, Project name for a job will be considerred when doing fairshare scheduling, i.e., as if user has submitted jobs using -G
enforceProjCheck
Enforces the check to see if the invoker of bsub is in the specifed group when the -P option is used
jobRunTimes
Run time for a job
dbDefaultIntval
Event table Job default interval
dbHjobCountIntval
Event table Job Host Count
dbQjobCountIntval
Event table Job Queue Count
dbUjobCountIntval
Event table Job User Count
dbJobResUsageIntval
Event table Job Resource Interval
dbLoadIntval
Event table Resource Load Interval
dbJobInfoIntval
Event table Job Info
jobDepLastSub
Used with job dependency scheduling
maxJobNameDep
Used with job dependency scheduling, deprecated
dbSelectLoad
select resources to be logged
jobSynJgrp
job synchronizes its group status
pjobSpoolDir
The batch jobs' temporary output directory
maxUserPriority
Maximal job priority defined for all users
jobPriorityValue
Job priority is increased by the system dynamically based on waiting time
jobPriorityTime
Waiting time to increase Job priority by the system dynamically
enableAutoAdjust
Enable internal statistical adjustment
autoAdjustAtNumPend
Start to autoadjust when the user has this number of pending jobs
autoAdjustAtPercent
If this number of jobs has been visited skip the user
sharedResourceUpdFactor
Static shared resource update interval for the cluster
scheRawLoad
schedule job based on raw load info
jobAttaDir
The batch jobs' external storage for attached data
maxJobMsgNum
Maximum message number for each job
maxJobAttaSize
Maximum attached data size to be transferred for each message
mbdRefreshTime
The life time of a child MBD to serve queries in the MT way
updJobRusageInterval
The interval of the execution cluster updating the job's resource usage
sysMapAcct
The account to which all windows workgroup users are to be mapped
preExecDelay
dispatch delay internal
updEventUpdateInterval
The interval of updating duplicated logging info
resourceReservePerSlot
Resources are reserved for parallel jobs on a per-slot basis
maxJobId
The Maximum JobId defined in the system
preemptResourceList
The list of preemption resources
preemptionWaitTime
The time for preemption wait
maxAcctArchiveNum
Max number of Acct files
acctArchiveInDays
Mbatchd Archive Interval
acctArchiveInSize
Mbatchd Archive threshold
committedRunTimeFactor
Committed run time weighting factor
enableHistRunTime
Enable the use of historical run time in the calculation of fairshare scheduling priority, Disable the use of historical run time in the calculation of fairshare scheduling priority
nqsUpdateInterval
NQS resource usage update interval
mcbOlmReclaimTimeDelay
Open lease reclaim time
chunkJobDuration
Enable chunk job dispatch for jobs with CPU limit or run limits
sessionInterval
The interval for scheduling jobs by scheduler daemon
publishReasonJobNum
The number of jobs per user per queue whose pending reason is published at the PEND_REASON_UPDATE_INTERVAL interval
publishReasonInterval
The interval for publishing job pending reason by scheduler daemon
publishReason4AllJobInterval
Interval (in seconds) of pending reason publish for all jobs
mcUpdPendingReasonInterval
MC pending reason update interval (0 means no updates)
mcUpdPendingReasonPkgSize
MC pending reason update package size (0 means no limit)
noPreemptRunTime
No preemption run time
noPreemptFinishTime
No preemption finish time
acctArchiveAt
Mbatchd Archive Time
absoluteRunLimit
Absolute run limit for job
lsbExitRateDuration
The job exit rate duration
lsbTriggerDuration
The duration to trigger eadmin
maxJobinfoQueryPeriod
Maximum time for job information query commands (for example,with bjobs) to wait
jobSubRetryInterval
Job submission retry interval
pendingJobThreshold
System wide max pending jobs
maxConcurrentJobQuery
Maximum concurrent job query
minSwitchPeriod
Minimal event switch period
condensePendingReasons
Condense pending reasons enabled
slotBasedParallelSched
Schedule Parallel jobs based on slots instead of CPUs
disableUserJobMovement
Job position control by admin
detectIdleJobAfter
Detect idle jobs after
useSymbolPriority
Use symbolic when specifing priority of session scheduler jobs
JobPriorityRound
Priority rounding for session scheduler jobs
priorityMapping
Priority rounding for session scheduler jobs
maxInfoDirs
Maximum number of info directories
minMbdRefreshTime
The minimum period of a child MBD to serve queries in the MT way
enableStopAskingLicenses2LS
Stop asking license to LS not due to lack license
mbdQueryCPUs
MBD child query processes will only run on the following CPUs
defaultApp
The default application profile assigned to jobs
enableStream
Streaming of lsbatch data is enabled
streamFile
File to which lsbatch data is streamed
streamSize
File size in MB to which lsbatch data is streamed
syncUpHostStatusWithLIM
Sync up host status with master LIM is enabled
defaultSLA
EGO Enabled SLA scheduling is enabled
slaTimer
EGO Enabled SLA scheduling timer period
mbdEgoTtl
EGO Enabled SLA scheduling time to live
mbdEgoConnTimeout
EGO Enabled SLA scheduling connection timeout
mbdEgoReadTimeout
EGO Enabled SLA scheduling read timeout
mbdUseEgoMXJ
EGO Enabled SLA scheduling use MXJ flag
mbdEgoReclaimByQueue
EGO Enabled SLA scheduling reclaim by queue
defaultSLAvelocity
EGO Enabled SLA scheduling default velocity
exitRateTypes
Type of host exit rate exception handling types: EXIT_RATE_TYPE
globalJobExitRate
Type of host exit rate exception handling types: GLOBAL_EXIT_RATE
enableJobExitRatePerSlot
Type of host exit rate exception handling types ENABLE_EXIT_RATE_PER_SLOT
enableMetric
Performance metrics monitor is enabled flag
schMetricsSample
Performance metrics monitor sample period flag
maxApsValue
used to bound: (1) factors, (2) weights, and (3) APS values
newjobRefresh
Child mbatchd gets updated information about new jobs from the parent mbatchd
preemptJobType
Job type to preempt, PREEMPT_JOBTYPE_BACKFILL, PREEMPT_JOBTYPE_EXCLUSIVE
defaultJgrp
The default job group assigned to jobs
jobRunlimitRatio
Max ratio between run limit and runtime estimation
jobIncludePostproc
Enable the post-execution processing of the job to be included as part of the job flag
jobPostprocTimeout
Timeout of post-execution processing
sschedUpdateSummaryInterval
The interval, in seconds, for updating the session scheduler status summary
sschedUpdateSummaryByTask
The number of completed tasks for updating the session scheduler status summary
sschedRequeueLimit
The maximum number of times a task can be requeued via requeue exit values
sschedRetryLimit
The maximum number of times a task can be retried after a dispatch error
sschedMaxTasks
The maximum number of tasks that can be submitted in one session
sschedMaxRuntime
The maximum run time of a single task
sschedAcctDir
The output directory for task accounting files
jgrpAutoDel
If TRUE enable the job group automatic deletion functionality (default is FALSE).
maxJobPreempt
Maximum number of job preempted times.
maxJobRequeue
Maximum number of job re-queue times.
noPreemptRunTimePercent
No preempt run time percent.
maxStreamFileNum
Maximum number of backup stream.utc files.
privilegedUserForceBkill
If enforced only admin can use the bkill -r option.
intersectCandidateHosts
Jobs run on only on hosts belonging to the intersection of the queue the job was submitted to, advance reservation hosts, and any hosts specified by bsub -m at the time of submission.
enforceOneUGLimit
Enforces the limitations of a single specified user group.
logRuntimeESTExceeded
Logging of runtime estimation events.
computeUnitTypes
Compute unit types.
fairAdjustFactor
Fairshare adjustment weighting factor.
noPreemptFinishTimePercent
No preempt finish time percent.
slotReserveQueueLimit
The reservation request being within JL/U.
maxJobPercentagePerSession
Job accept limit percentage.
useSuspSlots
The low priority job will use the slots freed by preempted jobs.
RETURN VALUES
pointer:parameterInfo
The function was a success, returns a pointer to a parameterInfo structure.
char:NULL
Function failed.
ERRORS
If the function fails, lsberrno is set to indicate the error.
SEE ALSO
Related APIs
none
Equivalent line command
none
Files
$LSB_CONFDIR/cluster_name/lsb.params
Platform Computing Inc.
www.platform.com |
Knowledge Center Contents Previous Next |