1 .TH STRESSAPPTEST 1 2009-10-20
3 stressapptest \- stress test application for simulating high load situations
8 This manual page documents briefly the
12 \fBstressapptest\fP (unix name for Stressful Application Test) is a program that
13 tries to maximize randomized traffic to memory from processor and I/O, with the
14 intent of creating a realistic high load situation in order to test the existing
15 hardware devices in a computer.
17 This program supports the following options:
21 Run in degraded mode on incompatible systems.
25 Number of memory CPU stress threads to run.
29 Add a direct write disk thread with block device (or file) 'device'.
33 Add a disk thread with tempfile 'filename'.
37 Don't result check each transaction.
41 Number of memory invert threads to run.
45 Log output to file 'logfile'.
49 Number of memory copy threads to run.
53 Megabytes of RAM to test.
57 Add a network thread connecting to system at 'ipaddr'.
61 Size in bytes of memory chunks.
65 Number of seconds to run.
69 Verbosity (0-20), default is 8.
73 Use more CPU-stressful memory copy.
76 .B \-\-blocks\-per\-segment <number>
77 Number of blocks to read/write per segment per iteration (\-d).
80 .B \-\-cache\-size <size>
81 Size of disk cache (\-d).
84 .B \-\-cc_inc_count <number>
85 Number of times to increment the cacheline's member.
88 .B \-\-cc_line_count <number>
89 Number of cache line sized datastructures to allocate for the cache coherency
93 .B \-\-cc_line_size <number>
94 Size of cache line to use as the basis for cache coherency test data
99 Do the cache coherency testing.
103 Write/wipe disk partition (\-d).
106 .B \-\-filesize <size>
107 Size of disk IO tempfiles.
111 Find locations to do disk IO automatically.
115 Inject false errors to test error handling.
118 .B \-\-force_errors_like_crazy
119 Inject a lot of false errors to test error handling.
123 Run threads that listen for incoming net connections.
127 Choose memory regions associated with each CPU to be tested by that CPU.
130 .B \-\-max_errors <number>
131 Exit early after finding specified number of errors.
135 Only do ECC error polling, no stress load.
139 Run without checking for ECC or other errors.
142 .B \-\-paddr_base <address>
143 Allocate memory starting from this address.
146 .B \-\-pause_delay <seconds>
147 Delay (in seconds) between power spikes.
150 .B \-\-pause_duration <seconds>
151 Duration (in seconds) of each pause.
154 .B \-\-random-threads <number>
155 Number of random threads for each disk write thread (\-d).
158 .B \-\-read-block-size <size>
159 Size of block for reading (\-d).
162 .B \-\-read-threshold <time>
163 Maximum time (in us) a block read should take (\-d).
166 .B \-\-remote_numa <time>
167 Choose memory regions not associated with each CPU to be tested by that CPU.
170 .B \-\-segment-size <size>
171 Size of segments to split disk into (\-d).
174 .B \-\-stop_on_errors
175 Stop after finding the first error.
178 .B \-\-write-block-size <size>
179 Size of block for writing (\-d). If not defined, the size of block for writing
180 will be defined as the size of block for reading.
183 .B \-\-write-threshold <time>
184 Maximum time (in us) a block write should take (\-d).
187 .BR http://code.google.com/p/stressapptest/
189 stressapptest was written by Nick Sanders and Rapahel Menderico (Google Inc).
191 This manual page was written by Michael Prokop <mika@debian.org>
192 for the Debian project (and may be used by others).