1sreport(1) Slurm Commands sreport(1)
2
6 sreport - Generate reports from the slurm accounting data.
7
10 sreport [OPTIONS...] [COMMAND...]
11
14 sreport is used to generate reports of job usage and cluster utiliza‐
15 tion for Slurm jobs saved to the Slurm Database, slurmdbd.
16
19 -a, --all_clusters
20 Use all clusters instead of only the cluster from which the com‐
21 mand was executed.
22 --federation
24 Generate a report for the federation if a member of one.
25 -h, --help
27 Print a help message describing the usage of sreport.
28 --local
30 Generate a report for the local cluster, even if part of a fed‐
31 eration. Overrides --federation
32 -M, --cluster
34 The cluster(s) to generate reports for. Default is local clus‐
35 ter, unless the local cluster is currently part of a federation
36 and in that case generate a report for all clusters in the cur‐
37 rent federation. If the clusters included in a federation vary
38 through time, use this option to identify the clusters to be in‐
39 cluded in report. Implies --local.
40 -n, --noheader
42 Don't display header when listing results.
43 -p, --parsable
45 Output will be '|' delimited with a '|' at the end.
46 -P, --parsable2
48 Output will be '|' delimited without a '|' at the end.
49 -Q, --quiet
51 Print no warning or informational messages, only error messages.
52 -t <format>
54 Specify the output time format. Time format options are case in‐
55 sensitive and may be abbreviated. The default format is Minutes.
56 Supported time format options are listed in the time command
57 section below.
58 -T, --tres <tres_names>
60 Trackable resource (TRES) to report values for. By default CPU
61 resource use is reported (except for reservation reports. All
62 TRES types reserved by a reservation will be shown by default
63 unless overridden with this option). Multiple TRES names may be
64 separated using a comma separated list for all reports except
65 the job reports, which can only support a single TRES name, or
66 "ALL" for all TRES.
67 The "Reported" Billing TRES is calculated from the largest
69 Billing TRES of each node multiplied by the time frame. For ex‐
70 ample, if a node is part of multiple partitions and each has a
71 different TRESBillingWeights defined the Billing TRES for the
72 node will be the highest of the partitions. If TRESBilling‐
73 Weights is not defined on any partition for a node then the
74 Billing TRES will be equal to the number of CPUs on the node.
75 TRES node usage is no longer reported in percent format or in
77 Cluster Utilization. Please use TRES CPU instead. The main is‐
78 sue with using node is in most configurations multiple jobs are
79 able to run on the same node. This makes TRES node accounting
80 count the same node multiple times in the same period. In ex‐
81 clusive node configurations, CPU accounting returns the same us‐
82 age node accounting would.
83 -v, --verbose
86 Print detailed event logging.
87 -V , --version
89 Print version information and exit.
90
93 <keyword> may be omitted from the execute line and sreport will
94 execute in interactive mode. sreport will process commands as
95 entered until explicitly terminated.
96 exit Terminate the execution of sreport. Identical to the quit com‐
99 mand.
100 help Display a description of sreport options and commands.
103 nonparsable
106 Return output to normal after parsable or parsable2 has been
107 set.
108 parsable
111 Output will be | delimited with an ending '|'.
112 parsable2
115 Output will be | delimited without an ending '|'.
116 quiet Print no warning or informational messages, only fatal error
119 messages.
120 quit Terminate the execution of sreport. Identical to the exit com‐
123 mand.
124 time <time_format>
127 Specify the output time format. Time format options are case in‐
128 sensitive and may be abbreviated. The default format is Minutes.
129 Supported time format options include:
130 SecPer Seconds/Percentage of Total
133 MinPer Minutes/Percentage of Total
135 HourPer Hours/Percentage of Total
137 Seconds Seconds
139 Minutes Minutes
141 Hours Hours
143 Percent Percentage of Total
145 verbose
148 Enable detailed event logging.
149 version
152 Display the sreport version number.
153 !! Repeat the last command executed.
156
159 Valid report types are:
160 cluster <REPORT> <OPTIONS>
161 job <REPORT> <OPTIONS>
163 reservation <REPORT> <OPTIONS>
165 user <REPORT> <OPTIONS>
167 <REPORT> options for each type include:
170 cluster - AccountUtilizationByUser, UserUtilizationByAccount,
172 UserUtilizationByWckey, Utilization, WCKeyUtilizationByUser
173 job - SizesByAccount, SizesByAccountAndWcKey, SizesByWckey
175 reservation - Utilization
177 user - TopUsage
179 NOTE: If OverSubscribe is configured to FORCE or YES in your slurm.conf
182 and the system is not configured to use preemption (PreemptMode=OFF)
183 accounting can easily grow to values greater than the actual utiliza‐
184 tion. It may be common on such systems to get error messages in the
185 slurmdbd log stating: "We have more allocated time than is possible."
186
189 cluster AccountUtilizationByUser
190 This report will display account utilization as it appears on
191 the hierarchical tree. Starting with the specified account or
192 the root account by default this report will list the underlying
193 usage with a sum on each level. Use the 'tree' option to span
194 the tree for better visibility.
195 NOTE: Idle reservation time will be split evenly among ac‐
197 counts/users given access to it. When a reservation is assigned
198 to whole accounts, the time will be counted in the association
199 for the accounts, not the user associations in the accounts. In
200 this case, the usage of a parent account can be larger than the
201 sum of its children.
202 cluster UserUtilizationByAccount
204 This report will display users by account in order of utiliza‐
205 tion without grouping multiple accounts by user into one, but
206 displaying them on separate lines.
207 cluster UserUtilizationByWCKey
209 This report will display users by wckey in order of utilization
210 without grouping multiple wckey by user into one, but displaying
211 them on separate lines.
212 cluster Utilization
214 This report will display total usage divided amongst Allocated,
215 Down, Planned Down, Idle, and Reserved time for selected clus‐
216 ters.
217 Allocated Time that nodes were in use with active jobs or an
219 active reservation. This does not include reserva‐
220 tions created with the MAINT or IGNORE_JOBS flags.
221 Down Time that nodes were marked as Down or time that
223 slurmctld was not responding (assuming TrackSlurm‐
224 ctldDown is set in slurmdbd.conf).
225 Idle Time where nodes had no active jobs or reserva‐
227 tions.
228 Overcommited Time of eligible jobs waiting in the queue over the
230 Reserved time. Unlike Reserved, this has no limit.
231 It is typically useful to determine whether your
232 system is overloaded and by how much.
233 PlannedDown Time that nodes were in use by a reservation cre‐
235 ated with the MAINT flag but not the IGNORE_JOBS
236 flag.
237 Reserved Time that a node spent idle with eligible jobs in
239 the queue that were unable to start due to time or
240 size constraints. The maximum time is limited to
241 the amount of time possible on the system. If this
242 value is not of importance for you then the number
243 can be grouped with idle time.
244 TresCount Configured number of TRES' on the reported nodes.
246 See also the TresName field.
247 Note: Reservations created with the IGNORE_JOBS flag are not
249 tracked in the Cluster Utilization report due to the fact that
250 allowing any current/active jobs to continue to run in the
251 reservation introduces the possibility for them to be accounted
252 for incorrectly.
253 Note: The default view for the "Cluster Utilization" report in‐
255 cludes the following fields: Cluster, Allocated, Down, Planned‐
256 Down, Idle, Reserved, Reported. You can include additional
257 fields like OverCommitted and TresCount fields with the Format
258 option. The TresName will also be included if using the -T,
259 --tres <tres_names> option.
260 cluster WCKeyUtilizationByUser
262 This report will display wckey utilization sorted by WCKey name
263 for each user on each cluster.
264 job SizesByAccount
267 This report will display the amount of time used for job ranges
268 specified by the 'grouping=' option. Only a single level in the
269 tree is displayed defaulting to the root dir. If you specify
270 other accounts with the 'account=' option sreport will use those
271 accounts as the root account and you will receive the aggregated
272 totals of each listed account plus their sub accounts.
273 job SizesByAccountAndWckey
275 This report is very similar to SizesByAccount with the differ‐
276 ence being each account is pair with wckeys so the identifier is
277 account:wckey instead of just account so there will most likely
278 be multiple accounts listed depending on the number of wckeys
279 used.
280 job SizesByWckey
282 This report will display the amount of time for each wckey for
283 job ranges specified by the 'grouping=' option.
284 reservation Utilization
287 This report will display total usage for reservations on the
288 systems. Note: Time requests on this report will not truncate
289 the time the reservation used, only the reservations that ran at
290 any time during the period requested.
291 user TopUsage
294 Displays the top users on a cluster, i.e. users with the highest
295 CPUTime. Use the group option to group accounts together. The
296 default is to have a different line for each user account combi‐
297 nation.
298 NOTE: TopUsage is always based on CPUTime. The -T, --tres option
300 will show the TRES usage of the top users, but the users are al‐
301 ways sorted by their CPUTime, not by the specified TRES.
302 Each report type has various options...
305 OPTIONS FOR ALL REPORT TYPES
308 All_Clusters
311 Use all monitored clusters. Default is local cluster.
312 Clusters=<OPT>
314 List of clusters to include in report. Default is local clus‐
315 ter.
316 End=<OPT>
319 Period ending for report. Default is 23:59:59 of previous day.
320 Valid time formats are...
321 HH:MM[:SS] [AM|PM]
323 MMDD[YY] or MM/DD[/YY] or MM.DD[.YY]
324 MM/DD[/YY]-HH:MM[:SS]
325 YYYY-MM-DD[THH:MM[:SS]]
326 now[{+|-}count[seconds(default)|minutes|hours|days|weeks]]
327 Format=<OPT>
330 Comma separated list of fields to display in report.
331 When using the format option for listing various fields you can
333 put a %NUMBER afterwards to specify how many characters should
334 be printed.
335 e.g. format=name%30 will print 30 characters of field name right
337 justified. A -30 will print 30 characters left justified.
338 Start=<OPT>
341 Period start for report. Default is 00:00:00 of previous day.
342 Valid time formats are...
343 HH:MM[:SS] [AM|PM]
345 MMDD[YY] or MM/DD[/YY] or MM.DD[.YY]
346 MM/DD[/YY]-HH:MM[:SS]
347 YYYY-MM-DD[THH:MM[:SS]]
348 now[{+|-}count[seconds(default)|minutes|hours|days|weeks]]
349 OPTIONS SPECIFICALLY FOR CLUSTER REPORTS
351 Accounts=<OPT>
354 When used with the UserUtilizationByAccount, or AccountUtiliza‐
355 tionByUser, List of accounts to include in report. Default is
356 all.
357 Tree When used with the AccountUtilizationByUser report will span the
359 accounts as they are in the hierarchy.
360 Users=<OPT>
362 When used with any report other than Utilization, List of users
363 to include in report. Default is all.
364 Wckeys=<OPT>
366 When used with the UserUtilizationByWckey or WCKeyUtilization‐
367 ByUser, List of wckeys to include in report. Default is all.
368 OPTIONS SPECIFICALLY FOR JOB REPORTS
371 Accounts=<OPT>
374 List of accounts to use for the report. Default is all which
375 will show only one line corresponding to the totals of all ac‐
376 counts in the hierarchy. This explanation does not apply when
377 ran with the FlatView or AcctAsParent options.
378 AcctAsParent
380 When used with the SizesbyAccount(*) will take specified ac‐
381 counts as parents and the next layer of accounts under those
382 specified will be displayed. Default is root if no specific Ac‐
383 counts are requested. When FlatView is used, this option is ig‐
384 nored.
385 FlatView
387 When used with the SizesbyAccount(*) will not group accounts in
388 a hierarchical level, but print each account where jobs ran on a
389 separate line without any hierarchy.
390 GID=<OPT>
392 List of group ids to include in report. Default is all.
393 Grouping=<OPT>
395 Comma separated list of size groupings. (e.g. 50,100,150 would
396 group job cpu count 1-49, 50-99, 100-149, > 150). grouping=in‐
397 dividual will result in a single column for each job size found.
398 Jobs=<OPT>
400 List of jobs/steps to include in report. Default is all.
401 Nodes=<OPT>
403 Only show jobs that ran on these nodes. Default is all.
404 Partitions=<OPT>
406 List of partitions jobs ran on to include in report. Default is
407 all.
408 PrintJobCount
410 When used with the Sizes report will print number of jobs ran
411 instead of time used.
412 Users=<OPT>
414 List of users jobs to include in report. Default is all.
415 Wckeys=<OPT>
417 List of wckeys to use for the report. Default is all. The
418 SizesbyWckey report all users summed together. If you want only
419 certain users specify them with the Users= option.
420 OPTIONS SPECIFICALLY FOR RESERVATION REPORTS
423 Names=<OPT>
425 List of reservations to use for the report. Default is all.
426 Nodes=<OPT>
428 Only show reservations that used these nodes. Default is all.
429 OPTIONS SPECIFICALLY FOR USER REPORTS
433 Accounts=<OPT>
436 List of accounts to use for the report. Default is all.
437 Group Group all accounts together for each user. Default is a sepa‐
439 rate entry for each user and account reference.
440 TopCount=<OPT>
442 Used in the TopUsage report. Change the number of users dis‐
443 played. Default is 10.
444 Users=<OPT>
446 List of users jobs to include in report. Default is all.
447
450 FORMAT OPTIONS FOR CLUSTER REPORTS
451 AccountUtilizationByUser:
452 UserUtilizationByAccount:
453 Accounts, Cluster, Login, Proper, TresCount, Used
454 UserUtilizationByWckey:
456 WCKeyUtilizationByUser:
457 Cluster, Login, Proper, TresCount, Used, Wckey
458 Utilization:
460 Allocated, Cluster, Down, Idle, Overcommitted, PlannedDown, Re‐
461 ported, Reserved, TresCount, TresName
462 FORMAT OPTIONS FOR JOB REPORTS
465 SizesByAccount, SizesByAccountAndWckey:
466 Account, Cluster
467 SizesByWckey:
469 Wckey, Cluster
470 FORMAT OPTIONS FOR RESERVATION REPORTS
473 Utilization:
474 Allocated, Associations, Cluster, End, Flags, Idle, Name, Nodes,
475 ReservationId, Start, TotalTime, TresCount, TresName, TresTime
476 FORMAT OPTIONS FOR USER REPORTS
479 TopUsage:
480 Account, Cluster, Login, Proper, Used
481 All commands and options are case-insensitive.
484
487 Executing sreport sends a remote procedure call to slurmdbd. If enough
488 calls from sreport or other Slurm client commands that send remote pro‐
489 cedure calls to the slurmdbd daemon come in at once, it can result in a
490 degradation of performance of the slurmdbd daemon, possibly resulting
491 in a denial of service.
492 Do not run sreport or other Slurm client commands that send remote pro‐
494 cedure calls to slurmdbd from loops in shell scripts or other programs.
495 Ensure that programs limit calls to sreport to the minimum necessary
496 for the information you are trying to gather.
497
500 Some sreport options may be set via environment variables. These envi‐
501 ronment variables, along with their corresponding options, are listed
502 below. (Note: commandline options will always override these settings)
503 SREPORT_CLUSTER Same as -M, --cluster
505 SREPORT_FEDERATION Same as --federation
507 SREPORT_LOCAL Same as --local
509 SREPORT_TRES Same as -t, --tres
511 SLURM_CONF The location of the Slurm configuration file.
513
516 Report number of jobs per account according to different job size bins:
517 $ sreport job sizesbyaccount
519 Report cluster utilization:
522 $ sreport cluster utilization
524 Report top usage:
527 $ sreport user top
529 Report number of jobs by user gore1 within the environ account:
532 $ sreport job sizesbyaccount All_Clusters users=gore1 account=environ PrintJobCount
534 Report cluster account utilization with the specified fields during the
537 specified 24 hour day of February 23, 2009, by user gore1:
538 $ sreport cluster AccountUtilizationByUser cluster=zeus user=gore1 start=2/23/09 end=2/24/09 format=Accounts,Cluster,TresCount,Login,Proper,Used
540 Report cluster account utilization by user in the LC account on cluster
543 zeus:
544 $ sreport cluster AccountUtilizationByUser cluster=zeus accounts=lc start=2/23/09 end=2/24/09
546 Report top usage in percent of the lc account during the specified
549 week:
550 $ sreport user topusage start=2/16/09 end=2/23/09 -t percent account=lc
552
555 Copyright (C) 2009-2010 Lawrence Livermore National Security. Produced
556 at Lawrence Livermore National Laboratory (cf, DISCLAIMER).
557 Copyright (C) 2010-2015 SchedMD LLC.
558 This file is part of Slurm, a resource management program. For de‐
560 tails, see <https://slurm.schedmd.com/>.
561 Slurm is free software; you can redistribute it and/or modify it under
563 the terms of the GNU General Public License as published by the Free
564 Software Foundation; either version 2 of the License, or (at your op‐
565 tion) any later version.
566 Slurm is distributed in the hope that it will be useful, but WITHOUT
568 ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
569 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
570 for more details.
571
574 sacct(1), slurmdbd(8)
575February 2021 Slurm Commands sreport(1)