qacct - report and account for Grid Engine usage
qacct [-ar [ar_id]] [-A
Account] [-b BeginTime] [-d Days]
[-e EndTime] [-g [GroupId|GroupName]]
[-h [HostName]] [-help] [-j
[job_id|job_name|pattern]] [-l
attr=val,...] [-m] [-o [Owner]]
[-u [Owner]] [-pe [PEname]] [-q
[wc_queue]] [-slots [SlotNumber]] [-t
task_id_range_list] [-P [Project]] [-D
[Department]] [-f AcctFileName] [-E]
The qacct utility scans the accounting data file (see
accounting(5)) and produces a summary of information for wall-clock
time, CPU time, and system time for the categories of host name, queue name,
group name, owner name, job name, job ID, and for the queues meeting the
resource requirements as specified with the -l switch. Combinations
of each category are permitted. Alternatively, all or specific jobs can be
listed with the -j switch. For example the search criteria could
include summarizing for a queue and an owner, but not for two queues in the
same request.
Note that results including parallel jobs depend on the value of
accounting_summary in the parallel environment definition (see
sge_pe(5)). If it is false, there is an accounting record for slave
tasks as well as master tasks so that, for example, there is a contribution
to CPU time from each task. Note that the slot count reported for each task
is the total for the job, not that allocated to the slave.
- -ar [ar_id]
- The ID of the advance reservation for which usage is summarized. If
ar_id is not given, accounting data are listed for each advance
reservation separately.
- -A
Account
- The account for jobs to be summarized.
- -b BeginTime
- The earliest start time for jobs to be summarized, in the format
[[CC]YY]MMDDhhmm[.SS]. See also the -d
option. If CC is not specified, a YY of <70 means
20YY. If the -E option is specified, this specifies the
earliest job end time instead.
- -d Days
- The number of days for which to summarize and print accounting
information. If used together with the -b BeginTime option
(see above), jobs started (or, with the -E option, finished) within
BeginTime to BeginTime+Days are counted. If used
together with -e EndTime (see below) option, selection is
from EndTime-Days. With neither, consider the last
Days.
- -e EndTime
- The latest start time for jobs to be summarized, in the format
[[CC]YY]MMDDhhmm[.SS]. See also the -d
option. If CC is not specified, a YY of <70 means
20YY. If the -E option is specified, this specifies the
latest job end time instead.
- [-f AcctFileName]
- The accounting file to be used. "-" means use standard input. If
omitted, the system default accounting file is processed.
- -g
[GroupId|GroupName]
- The numeric system group id or the group alphanumeric name of the job
owners to be included in the accounting. If
GroupId/GroupName is omitted, all groups are accounted.
- -h [HostName]
- The case-insensitive name of the host upon which accounting information is
requested. If the name is omitted, totals for each host are listed
separately.
- -help
- Display help information for the qacct command.
- -j
[job_id|job_name|pattern]
- The name, an expression for matching names, or ID of the job during
execution for which accounting information is printed. If neither a name
nor an ID is given, all jobs are listed.
This option changes the output format of qacct. If activated, CPU
times are no longer accumulated but "raw" per-job accounting
information is printed in a formatted form instead. CPU time (seconds),
integrated memory usage (GB seconds), total i/o (bytes), i/o wait time
(seconds), and maximum virtual memory (bytes).
- -l
attr=val,...
- A resource requirement specification which must be met by the queues in
which the jobs being accounted were executing. The resource request is
very similar to the one described in qsub(1).
- -m
- Ignore slave tasks of parallel jobs. Only relevant when when
accounting_summary is false, and probably only useful with
-j to prune the output.
- -o [Owner] | -u
[Owner]
- The name of the owner of the jobs for which accounting statistics are
assembled. If the optional Owner argument is omitted, a listing of
the accounting statistics of all job owners present in the accounting file
is produced. -u and -o are synonyms.
- -pe
[PEname]
- The name of the parallel environment for which usage is summarized. If
PEname is not given, accounting data are listed for each parallel
environment separately.
- -q
[wc_queue_name]
- An expression for queues for which usage is summarized. All queue
instances matching the expression will be listed. If no expression is
specified, a cluster queue summary will be given.
- -slots
[SlotNumber]
- The number of queue slots for which usage is summarized. If
SlotNumber is not given, accounting data are listed for each number
of queue slots separately.
- -t
task_id_range_list
- Only available together with the -j option described above.
The -t switch specifies the array job task range, for
which accounting information should be printed. Syntax and semantics of
task_id_range_list are identical to that one described under the
-t option to qsub(1). Please see there also for further
information on array jobs.
- -P [Project]
- The name of the project for which usage is summarized. If Project
is not given, accounting data are listed for each relevant project
separately.
- -D
[Department]
- The name of the department for which usage is summarized. If
Department is not given, accounting data are listed for each
relevant department separately.
- -E
- When selecting times with -b, -e, -d, use the job end
time rather than the start time to select. This may be useful because the
accounting record is written at the job end.
- SGE_ROOT
- Specifies the location of the Grid Engine standard configuration
files.
- SGE_CELL
- If set, specifies the default Grid Engine cell. To address a Grid Engine
cell qacct uses (in the order of precedence):
The name of the cell specified in the environment
variable SGE_CELL, if it is set.
The name of the default cell, i.e. default.
- SGE_DEBUG_LEVEL
- If set, specifies that debug information should be written to stderr. In
addition the level of detail in which debug information is generated is
defined.
- SGE_QMASTER_PORT
- If set, specifies the tcp port on which sge_qmaster(8) is expected
to listen for communication requests. Most installations will use a
services map entry for the service "sge_qmaster" instead to
define that port.
- SGE_EXECD_PORT
- If set, specifies the tcp port on which sge_execd(8) is expected to
listen for communication requests. Most installations will use a services
map entry for the service "sge_execd" instead to define that
port.
<sge_root>/<cell>/common/accounting
Grid Engine default accounting file
<xxqs_nam_sexx_root>/<cell>/common/act_qmaster
Grid Engine master host file
See sge_intro(1) for a full statement of rights and
permissions.