1ACCOUNTING(5) Grid Engine File Formats ACCOUNTING(5)
2
6 accounting - Grid Engine accounting file format
7
9 An accounting record is written to the Grid Engine accounting file for
10 each job having finished. The accounting file is processed by qacct(1)
11 to derive accounting statistics.
12
14 Each job is represented by a line in the accounting file. Empty lines
15 and lines which contain one character or less are ignored. Accounting
16 record entries are separated by colon (':') signs. The entries denote
17 in their order of appearance:
18 qname
20 Name of the cluster queue in which the job has run.
21 hostname
23 Name of the execution host.
24 group
26 The effective group id of the job owner when executing the job.
27 owner
29 Owner of the Grid Engine job.
30 job_name
32 Job name.
33 job_number
35 Job identifier - job number.
36 account
38 An account string as specified by the qsub(1) or qalter(1) -A option.
39 priority
41 Priority value assigned to the job corresponding to the priority
42 parameter in the queue configuration (see queue_conf(5)).
43 submission_time
45 Submission time (GMT unix time stamp).
46 start_time
48 Start time (GMT unix time stamp).
49 end_time
51 End time (GMT unix time stamp).
52 failed
54 Indicates the problem which occurred in case a job could not be started
55 on the execution host (e.g. because the owner of the job did not have a
56 valid account on that machine). If Grid Engine tries to start a job
57 multiple times, this may lead to multiple entries in the accounting
58 file corresponding to the same job ID.
59 exit_status
61 Exit status of the job script (or Grid Engine specific status in case
62 of certain error conditions). The exit status is determined by follow‐
63 ing the normal shell conventions. If the command terminates normally
64 the value of the command is its exit status. However, in the case that
65 the command exits abnormally, a value of 0200 (octal), 128 (decimal) is
66 added to the value of the command to make up the exit status.
67 For example: If a job dies through signal 9 (SIGKILL) then the
69 exit status becomes 128 + 9 = 137.
70 ru_wallclock
72 Difference between end_time and start_time (see above).
73 The remainder of the accounting entries follows the contents of the
75 standard UNIX rusage structure as described in getrusage(2). Depending
76 on the operating system where the job was executed some of the fields
77 may be 0. The following entries are provided:
78 ru_utime
80 ru_stime
81 ru_maxrss
82 ru_ixrss
83 ru_ismrss
84 ru_idrss
85 ru_isrss
86 ru_minflt
87 ru_majflt
88 ru_nswap
89 ru_inblock
90 ru_oublock
91 ru_msgsnd
92 ru_msgrcv
93 ru_nsignals
94 ru_nvcsw
95 ru_nivcsw
96 project
98 The project which was assigned to the job.
99 department
101 The department which was assigned to the job.
102 granted_pe
104 The parallel environment which was selected for that job.
105 slots
107 The number of slots which were dispatched to the job by the scheduler.
108 task_number
110 Array job task index number.
111 cpu
113 The cpu time usage in seconds.
114 mem
116 The integral memory usage in Gbytes cpu seconds.
117 io
119 The amount of data transferred in input/output operations.
120 category
122 A string specifying the job category.
123 iow
125 The io wait time in seconds.
126 pe_taskid
128 If this identifier is set the task was part of a parallel job and was
129 passed to Grid Engine via the qrsh -inherit interface.
130 maxvmem
132 The maximum vmem size in bytes.
133 arid
135 Advance reservation identifier. If the job used resources of an advance
136 reservation then this field contains a positive integer identifier oth‐
137 erwise the value is "0" .
138 ar_submission_time
140 If the job used resources of an advance reservation then this field
141 contains the submission time (GMT unix time stamp) of the advance
142 reservation, otherwise the value is "0" .
143
145 ge_intro(1), qacct(1), qalter(1), qsub(1), getrusage(2), queue_conf(5).
146
148 See ge_intro(1) for a full statement of rights and permissions.
149GE 6.2u5 $Date: 2008/12/23 16:51:59 $ ACCOUNTING(5)