1PMLOGREDUCE(1) General Commands Manual PMLOGREDUCE(1)
2
3
4
6 pmlogreduce - temporal reduction of Performance Co-Pilot archives
7
9 $PCP_BINADM_DIR/pmlogreduce [-z] [-A align] [-S starttime] [-s samples]
10 [-T endtime] [-t interval] [-v volsamples] [-Z timezone] input output
11
13 pmlogreduce reads one Performance Co-Pilot (PCP) archive identified by
14 input (this must be a PCP archive created by pmlogger(1), pmlogex‐
15 tract(1) or pmlogreduce(1)), and creates a temporally reduced PCP ar‐
16 chive in output. The data reduction involves statistical and temporal
17 reduction of samples with an output sampling interval defined by the -t
18 option in the output archive (independent of the sampling intervals in
19 the input archive), and is further controlled by other command line
20 arguments.
21
23 The command line options for pmlogreduce are as follows:
24
25 -A align
26 Specify a ``natural'' alignment of the output sample times;
27 refer to PCPIntro(1).
28
29 -S starttime
30 Define the start of a time window to restrict the samples
31 retrieved from the input archive; refer to PCPIntro(1).
32
33 -s samples
34 The argument samples defines the number of samples to be written
35 to output. If samples is 0 or -s is not specified, pmlogreduce
36 will sample until the end of the PCP archive, or the end of the
37 time window as specified by -T, whichever comes first. The -s
38 option will override the -T option if it occurs sooner.
39
40 -T endtime
41 Define the termination of a time window to restrict the samples
42 retrieved from the input archive; refer to PCPIntro(1).
43
44 -v volsamples
45 The output archive is potentially a multi-volume data set, and
46 the -v option causes pmlogreduce to start a new volume after
47 volsamples log records have been written to the output archive.
48
49 -t interval
50 Consecutive samples in the output archive will appear with a
51 time delta defined by interval; refer to PCPIntro(1). Note the
52 default value is 600 (seconds, i.e. 10 minutes).
53
54 -Z timezone
55 Use timezone when displaying the date and time, or interpreting
56 the -S and -T options. Timezone is in the format of the envi‐
57 ronment variable TZ as described in environ(5).
58
59 -z Use the local timezone of the host from the input archive when
60 displaying the date and time, or interpreting the -S and -T
61 options. The default is to initially use the timezone of the
62 local host.
63
65 The statistical and temporal reduction follows the following rules:
66
67 x. Consecutive records from input are read without interpolation, and
68 at most one output record is written for each interval, summarizing
69 the performance data over that period.
70
71 x. If the semantics of a metric indicates it is instantaneous or dis‐
72 crete then output value is computed as the arithmetic mean of the
73 observations (if any) over each interval.
74
75 x. If the semantics of a metric indicates it is a counter then the
76 following transformations are applied:
77 a) Metrics with 32-bit precision are promoted to 64-bit precision.
78 b) Any counter wrap (overflow) is noted, and appropriate adjust‐
79 ment made in the value of the metric over each interval. This
80 will be correct in the case of a single counter wrap, but will
81 silently underestimate in the case where more than one counter
82 wrap occurs between consecutive observations in the input ar‐
83 chive, and silently overestimate in the case where a counter is
84 reset occurs between consecutive observations in the input ar‐
85 chive; unfortunately these situations cannot be detected, but
86 are believed to be rare events for the sort of production moni‐
87 toring environments where pmlogreduce is most likely to be
88 deployed.
89
90 x. Any changes in instance domains, and indeed all metadata, is pre‐
91 served.
92
93 x. Any ``mark'' records in the input archive (as created by pmlogex‐
94 tract(1)) will be preserved in the output archive, so periods where
95 no data is available are maintained, and data interpolation will
96 not occur across these periods when the output archive is subse‐
97 quently processed with PCP applications.
98
100 For each of the input and output archives, several physical files are
101 used.
102 archive.meta
103 metadata (metric descriptions, instance domains, etc.) for
104 the archive log
105 archive.0 initial volume of metrics values (subsequent volumes have
106 suffixes 1, 2, ...)
107 archive.index
108 temporal index to support rapid random access to the other
109 files in the archive log.
110
112 Environment variables with the prefix PCP_ are used to parameterize the
113 file and directory names used by PCP. On each installation, the file
114 /etc/pcp.conf contains the local values for these variables. The
115 $PCP_CONF variable may be used to specify an alternative configuration
116 file, as described in pcp.conf(4).
117
119 PCPIntro(1), pmdumplog(1), pmlc(1), pmlogextract(1), pmlogger(1),
120 pcp.conf(4) and pcp.env(4).
121
123 All error conditions detected by pmlogreduce are reported on stderr
124 with textual (if sometimes terse) explanation.
125
126 Should the input archive be corrupted (this can happen if the pmlogger
127 instance writing the archive suddenly dies), then pmlogreduce will
128 detect and report the position of the corruption in the file, and any
129 subsequent information from the input archive will not be processed.
130
131 If any error is detected, pmlogreduce will exit with a non-zero status.
132
134 The preamble metrics (pmcd.pmlogger.archive, pmcd.pmlogger.host, and
135 pmcd.pmlogger.port), which are automatically recorded by pmlogger at
136 the start of the archive, may not be present in the archive output by
137 pmlogreduce. These metrics are only relevant while the archive is
138 being created, and have no significance once recording has finished.
139
140
141
142Performance Co-Pilot SGI PMLOGREDUCE(1)