In this benchmark, I did nothing with Laravel itself, I'm just want to see the different behaviors of the different optimization ways.
First of all, I'd like to put the conclusion here. The OPcache is a good choice to speed up your PHP project. From the two pics below, it seems like I didn't get the advantage of PHP PM, maybe it was limited by my Docker memory configuration. Anyway, the response latency of PHP PM is very impressive, 98% of requests are within 84ms.
Complete Requests per Second
Percentage of the Requests Served within a Certain Time (ms)
The first is round of this benchmark, I just pulled down the Laravel code from Github and run the cache commands.
The second round followed this post from Taylor Otwell. It will tell you about the how and why questions of Laravel benchmark.
For the last one, PPM, it's a concurrency framework based on ReactPHP. As an event-driven framework is different from the way of how I used to construct my code.
Benchmark command line:
ab -t 10 -c 10 http://test.vagrant/Default Settings
➜ ~ ab -t 10 -c 10 http://test.vagrant/
This is ApacheBench, Version 2.3 <$Revision: 1757674 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/
Benchmarking test.vagrant (be patient)
Finished 47 requests
Server Software: Apache/2.4.10
Server Hostname: test.vagrant
Server Port: 80
Document Path: /
Document Length: 2321 bytes
Concurrency Level: 10
Time taken for tests: 10.035 seconds
Complete requests: 47
Failed requests: 0
Total transferred: 157750 bytes
HTML transferred: 109087 bytes
Requests per second: 4.68 [#/sec] (mean)
Time per request: 2135.081 [ms] (mean)
Time per request: 213.508 [ms] (mean, across all concurrent requests)
Transfer rate: 15.35 [Kbytes/sec] received
Connection Times (ms)
min mean[+/-sd] median max
Connect: 0 0 0.0 0 0
Processing: 1324 1988 313.9 1973 3017
Waiting: 1324 1988 313.9 1973 3017
Total: 1324 1989 313.9 1973 3017
Percentage of the requests served within a certain time (ms)
50% 1949
66% 2136
75% 2202
80% 2258
90% 2342
95% 2395
98% 3017
99% 3017
100% 3017 (longest request)With OPcache optimized
➜ source ab -t 10 -c 10 http://test.vagrant/
This is ApacheBench, Version 2.3 <$Revision: 1757674 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/
Benchmarking test.vagrant (be patient)
Finished 678 requests
Server Software: Apache/2.4.10
Server Hostname: test.vagrant
Server Port: 80
Document Path: /
Document Length: 2321 bytes
Concurrency Level: 10
Time taken for tests: 10.020 seconds
Complete requests: 678
Failed requests: 0
Total transferred: 2275602 bytes
HTML transferred: 1573638 bytes
Requests per second: 67.66 [#/sec] (mean)
Time per request: 147.788 [ms] (mean)
Time per request: 14.779 [ms] (mean, across all concurrent requests)
Transfer rate: 221.78 [Kbytes/sec] received
Connection Times (ms)
min mean[+/-sd] median max
Connect: 0 0 0.1 0 1
Processing: 45 140 208.3 102 2097
Waiting: 45 140 208.4 102 2097
Total: 46 140 208.3 102 2097
Percentage of the requests served within a certain time (ms)
50% 102
66% 118
75% 129
80% 138
90% 159
95% 203
98% 1151
99% 1219
100% 2097 (longest request)PHP PM
➜ ~ ab -t 10 -c 10 http://test.vagrant/
This is ApacheBench, Version 2.3 <$Revision: 1757674 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/
Benchmarking test.vagrant (be patient)
Finished 836 requests
Server Software: nginx/1.12.2
Server Hostname: test.vagrant
Server Port: 80
Document Path: /
Document Length: 2321 bytes
Concurrency Level: 10
Time taken for tests: 10.001 seconds
Complete requests: 836
Failed requests: 0
Total transferred: 3343384 bytes
HTML transferred: 1940356 bytes
Requests per second: 83.59 [#/sec] (mean)
Time per request: 119.629 [ms] (mean)
Time per request: 11.963 [ms] (mean, across all concurrent requests)
Transfer rate: 326.47 [Kbytes/sec] received
Connection Times (ms)
min mean[+/-sd] median max
Connect: 0 0 0.1 0 1
Processing: 16 102 603.2 40 7371
Waiting: 16 101 603.2 40 7370
Total: 17 102 603.2 40 7371
Percentage of the requests served within a certain time (ms)
50% 40
66% 45
75% 50
80% 54
90% 62
95% 68
98% 84
99% 2341
100% 7371 (longest request)