1 | == micro-benchmark |
---|
2 | [[toc:]] |
---|
3 | |
---|
4 | === Repository |
---|
5 | |
---|
6 | [[https://bitbucket.org/certainty/micro-benchmark/overview]] |
---|
7 | |
---|
8 | |
---|
9 | === Authors |
---|
10 | [[/users/david-krentzlin|David Krentzlin]] |
---|
11 | |
---|
12 | |
---|
13 | === Introduction |
---|
14 | |
---|
15 | This egg provides means to benchmark your code in a simple manner. It features traditional benchmarking with microsecond |
---|
16 | resultion as well as a different approach that has been inspired by: [[https://github.com/evanphx/benchmark-ips]] |
---|
17 | |
---|
18 | |
---|
19 | |
---|
20 | === Examples |
---|
21 | |
---|
22 | <enscript highlight=scheme> |
---|
23 | (use micro-benchmark) |
---|
24 | |
---|
25 | ;; simply measure the runtime of the given fragment |
---|
26 | (benchmark-measure (sleep 2)) |
---|
27 | |
---|
28 | ;; run code 3 times and return results |
---|
29 | (parameterize ((current-benchmark-iterations 3)) |
---|
30 | (benchmark-run (sleep 1))) |
---|
31 | |
---|
32 | ;; find out how many iterations we can make per second |
---|
33 | (benchmark-ips (sleep 2)) |
---|
34 | </enscript> |
---|
35 | |
---|
36 | === API |
---|
37 | |
---|
38 | <parameter>current-benchmark-iterations</parameter> |
---|
39 | The amount of iterations that are used in '''benchmark-run'''. This defaults to 100. |
---|
40 | |
---|
41 | <syntax>(benchmark-measure ?code)</syntax> |
---|
42 | Runs the '''?code''' once and returns the runtime in microseconds. |
---|
43 | |
---|
44 | <syntax>(benchmark-run [iterations] ?code)</syntax> |
---|
45 | Runs the '''?code''' '''iterations''' times and returns an alist with the following keys: |
---|
46 | |
---|
47 | * min - the minimum runtime of all iterations in microseconds |
---|
48 | * max - the maximum runtime of all iterations in microseconds |
---|
49 | * mean - the average runtime of all iterations in microseconds |
---|
50 | * standard-deviation - the sample standard deviation for the given runtimes |
---|
51 | |
---|
52 | If '''iterations''' is not given then '''curren-benchmark-iterations''' is used. |
---|
53 | |
---|
54 | <syntax>(benchmark-ips [seconds] ?code)</syntax> |
---|
55 | Determines how many times one can run the given '''?code''' per second and returns an alist with the following keys: |
---|
56 | |
---|
57 | * mean - the mean amount iterations we can make per second |
---|
58 | * standard-deviation - the sample standard deviation for the given iterations |
---|
59 | |
---|
60 | If '''seconds''' is not given then it defaults to 5. |
---|
61 | |
---|
62 | === License |
---|
63 | This program is free software: you can redistribute it and/or modify |
---|
64 | it under the terms of the GNU General Public License as published by |
---|
65 | the Free Software Foundation, either version 3 of the License, or (at |
---|
66 | your option) any later version. |
---|
67 | |
---|
68 | This program is distributed in the hope that it will be useful, but |
---|
69 | WITHOUT ANY WARRANTY; without even the implied warranty of |
---|
70 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
---|
71 | General Public License for more details. |
---|
72 | |
---|
73 | A full copy of the GPL license can be found at |
---|
74 | <http://www.gnu.org/licenses/>. |
---|