1HYPERFINE(1) General Commands Manual HYPERFINE(1)
2
6 hyperfine - command-line benchmarking tool
7
9 hyperfine [-ihV] [-w warmupruns] [-r runs] [-p cmd...] [-c cmd] [-s
10 style] [cmd...]
11
13 A command-line benchmarking tool which includes:
14 * Statistical analysis across multiple runs
16 * Support for arbitrary shell commands
17 * Constant feedback about the benchmark progress and current es‐
18 timates
19 * Warmup runs can be executed before the actual benchmark
20 * Cache-clearing commands can be set up before each timing run
21 * Statistical outlier detection to detect interference from
22 other programs and caching effects
23 * Export results to various formats: CSV, JSON, Markdown, Asci‐
24 iDoc
25 * Parameterized benchmarks (e.g. vary the number of threads)
26
28 -w, --warmup warmupruns
29 Perform warmupruns (number) before the actual benchmark. This
31 can be used to fill (disk) caches for I/O-heavy programs.
32 -m, --min-runs minruns
34 Perform at least minruns (number) runs for each command. De‐
36 fault: 10.
37 -M, --max-runs maxruns
39 Perform at most maxruns (number) runs for each command. Default:
41 no limit.
42 -r, --runs runs
44 Perform exactly runs (number) runs for each command. If this op‐
46 tion is not specified, hyperfine automatically determines the
47 number of runs.
48 -p, --prepare cmd...
50 Execute cmd before each timing run. This is useful for clearing
52 disk caches, for example. The --prepare option can be specified
53 once for all commands or multiple times, once for each command.
54 In the latter case, each preparation command will be run prior
55 to the corresponding benchmark command.
56 -c, --cleanup cmd
58 Execute cmd after the completion of all benchmarking runs for
60 each individual command to be benchmarked. This is useful if the
61 commands to be benchmarked produce artifacts that need to be
62 cleaned up.
63 -P, --parameter-scan var min max
65 Perform benchmark runs for each value in the range min..max. Re‐
67 places the string '{var}' in each command by the current parame‐
68 ter value.
69 Example:
71 hyperfine -P threads 1 8 'make -j {threads}'
72 This performs benchmarks for 'make -j 1', 'make -j 2', ...,
74 'make -j 8'.
75 -D, --parameter-step-size delta
77 This argument requires --parameter-scan to be specified as well.
79 Traverse the range min..max in steps of delta.
80 Example:
82 hyperfine -P delay 0.3 0.7 -D 0.2 'sleep {delay}'
83 This performs benchmarks for 'sleep 0.3', 'sleep 0.5' and 'sleep
85 0.7'.
86 -L, --parameter-list var values
88 Perform benchmark runs for each value in the comma-separated
90 list of values. Replaces the string '{var}' in each command by
91 the current parameter value.
92 Example:
94 hyperfine -L compiler gcc,clang '{compiler} -O2 main.cpp'
95 This performs benchmarks for 'gcc -O2 main.cpp' and 'clang -O2
97 main.cpp'.
98 -s, --style type
100 Set output style type (default: auto). Set this to 'basic' to
102 disable output coloring and interactive elements. Set it to
103 'full' to enable all effects even if no interactive terminal was
104 detected. Set this to 'nocolor' to keep the interactive output
105 without any colors. Set this to 'color' to keep the colors with‐
106 out any interactive output. Set this to 'none' to disable all
107 the output of the tool.
108 -S, --shell shell
110 Set the shell to use for executing benchmarked commands.
112 -i, --ignore-failure
114 Ignore non-zero exit codes of the benchmarked programs.
116 -u, --time-unit unit
118 Set the time unit to be used. Default: second. Possible values:
120 millisecond, second.
121 --export-asciidoc file
123 Export the timing summary statistics as an AsciiDoc table to the
125 given file.
126 --export-csv file
128 Export the timing summary statistics as CSV to the given file.
130 If you need the timing results for each individual run, use the
131 JSON export format.
132 --export-json file
134 Export the timing summary statistics and timings of individual
136 runs as JSON to the given file.
137 --export-markdown file
139 Export the timing summary statistics as a Markdown table to the
141 given file.
142 --show-output
144 Print the stdout and stderr of the benchmark instead of sup‐
146 pressing it. This will increase the time it takes for benchmarks
147 to run, so it should only be used for debugging purposes or when
148 trying to benchmark output speed.
149 -n, --command-name name
151 Identify a command with the given name. Commands and names are
153 paired in the same order: the first command executed gets the
154 first name passed as option.
155 -h, --help
157 Print help message.
159 -V, --version
161 Show version information.
163
165 Basic benchmark of 'find . -name todo.txt':
166 hyperfine 'find . -name todo.txt'
167 Perform benchmarks for 'sleep 0.2' and 'sleep 3.2' with a minimum 5
169 runs each:
170 hyperfine --min-runs 5 'sleep 0.2' 'sleep 3.2'
171 Perform a benchmark of 'grep' with a warm disk cache by executing 3
173 runs up front that are not part of the measurement:
174 hyperfine --warmup 3 'grep -R TODO *'
175 Export the results of a parameter scan benchmark to a markdown table:
177 hyperfine --export-markdown output.md --parameter-scan time 1 5 'sleep {time}'
178
180 David Peter (sharkdp)
181 Source, bug tracker, and additional information can be found on GitHub
183 at: https://github.com/sharkdp/hyperfine
184 HYPERFINE(1)