Slide 1
Slide 1 text
PHP7.1 is fast(?)
Slide 2
Slide 2 text
name:
͍ͩ͠͡ΎΜ͍ͪ
(Ishida)
nick:
uzulla
(͏ͣΒ)
fun:
PHP and !
Slide 3
Slide 3 text
conclusion
7.1 is fast !
(but any evidence here)
Slide 4
Slide 4 text
need benchmark !!
Slide 5
Slide 5 text
by the way...
php can't to be standalone httpd. right?
(php -s is for only development use)
(btw, hhvm can be !)
If will benchmark with apache/nginx, that is unfair(???)
Slide 6
Slide 6 text
use "ReactPHP"!
(use libevent)
standalone httpd(and other), that written by PHP!
Slide 7
Slide 7 text
Test Enviroment
4 VPS (Vultr, $5/mo)
4 1Core
4 1GB mem
4 ubuntu 16.10
4 PHP7.1.2
Slide 8
Slide 8 text
Write git clone code.
4 httpd, written by php
4 https://github.com/uzulla/
ReactPHP_PostSupportSample
4 include monolog, twig(template engine.)
Slide 9
Slide 9 text
__DIR__.'/twig_cache',
]);
$access_log = new \Monolog\Logger('access');
$access_log->pushHandler(new \Monolog\Handler\StreamHandler(__DIR__.'/access.log', \Monolog\Logger::INFO));
// app
$app = function (\Uzulla\React\Http\Request $req,\React\Http\Response $res) use ($twig, $access_log) {
$path = $req->getPath();
$method = $req->getMethod();
$access_log->addInfo("{$method}\t{$path}");
$params = [
'params'=>$req->getParams(),
'query'=>$req->getQuery()
];
if($method === 'GET' && preg_match('|\A/\z|u', $path)){
$res->writeHead(200, ['Content-Type' => 'text/html']);
$res->end($twig->render('index.twig',$params));
}elseif($method === 'POST' && preg_match('|\A/\z|u', $path)){
$res->writeHead(200, ['Content-Type' => 'text/html']);
$res->end($twig->render('index.twig',$params));
}else{
$res->writeHead(404, ['Content-Type' => 'text/html']);
$res->end($twig->render('notfound.twig'));
}
};
// build reactor
$loop = React\EventLoop\Factory::create();
$socket = new React\Socket\Server($loop);
$http = new \Uzulla\React\Http\Server($socket);
$http->on('request', $app);
$socket->listen(8080);
echo "running...\n";
$loop->run();
Slide 10
Slide 10 text
$ git clone THAT
$ composer install
$ php boot.php
listen...
Slide 11
Slide 11 text
ab(apache bench)
4 ab -n 10000 -c 10 http://127.0.0.1:8080/
Slide 12
Slide 12 text
Requests per second:
3477.39 #/sec
Slide 13
Slide 13 text
okay, try more concurrent connection.
Slide 14
Slide 14 text
5000 concurrent
4 ab -n 10000 -c 5000 http://127.0.0.1:8080/
Slide 15
Slide 15 text
fail
4 socket: Too many open files (24)
Slide 16
Slide 16 text
ok, ab's limits.
try another tool.
apt install wrk
4 https://github.com/wg/wrk
Slide 17
Slide 17 text
wrk
4 wrk -c 5000 -d 5 http://127.0.0.1:8080/
Slide 18
Slide 18 text
Requests/sec
98.16
Slide 19
Slide 19 text
tooooo slow,
hmmm, concurrent overkill?
try 1000 concurrent (same as ab's test).
Slide 20
Slide 20 text
4 wrk -c 1000 -d 5 http://127.0.0.1:8080/
4 Requests/sec: 83.17
Slide 21
Slide 21 text
!
Slide 22
Slide 22 text
okey, wrk is dosen't work.
try with httperf !
4 https://github.com/httperf/httperf
Slide 23
Slide 23 text
httperf
4 httperf --port=8080 --server=127.0.0.1 --
rate=2800 --num-conns=10000 --hog
Slide 24
Slide 24 text
Request rate:
2799.6 req/s
Slide 25
Slide 25 text
seems good.
but, is this correct?
I need try more tools !
Slide 26
Slide 26 text
siege
4 siege -c 100 http://127.0.0.1:8080/
4 Transaction rate: 393.35 trans/sec
Slide 27
Slide 27 text
Untrustworthy...
(Of course, these bench mark tools are more
trustworthy than weird httpd.)
Slide 28
Slide 28 text
all results (?)
ab: 3477.3 #/sec
wrk: 98.1 req/sec
httperf: 2799.6 req/s
siege: 393.3 trans/sec
Slide 29
Slide 29 text
me:
I'm screwed
lestrrat ( a.k.a daisuke maki):
Why don't you write a bechmarker yourself ?
me:
what?
Slide 30
Slide 30 text
me:
OK, I'll write tools!
(at YESTERDAY 11:36 pm)
Slide 31
Slide 31 text
I wrote "benchmark.php" !! (ta daaa!)
Slide 32
Slide 32 text
$i; $i++ ){
$res = file_get_contents($target);
}
$used_time = (microtime(true) - $start_usec) ;
echo ceil( 1/( $used_time/$c ) );
echo "req/sec";
Slide 33
Slide 33 text
haha... yes, just kidding.
let's play.
Slide 34
Slide 34 text
(unreliable) result
4 3,731 req/sec
Slide 35
Slide 35 text
I can't believe this result.
(faster than ab ;lol)
it's not bad.
but ...
Slide 36
Slide 36 text
I thought, this is slow benchmark tool. !
I want more faster". but how?
I need concurrent.!
but, this is php.
(curl_multi ? yes...but... )
OK! WE NEED FORK! !
Slide 37
Slide 37 text
$x;$x++){
$pid = pcntl_fork();
if ($pid == -1) { die("fork failed");}
else if ($pid) { $pids[]=$pid; }
else {
for($i=0;$c_>$i;$i++){ $res = file_get_contents($target); } exit;
}
}
foreach($pids as $pid){ pcntl_waitpid($pid, $status); }
$used_time = (microtime(true) - $start_usec) ;
echo ceil( 1/( $used_time/$c ) ), "req/sec";
Slide 38
Slide 38 text
2881req/sec
(no fork ver 3,731 req/sec)
Slide 39
Slide 39 text
no no, more slow !!! !
Slide 40
Slide 40 text
!
need more core!!!
Slide 41
Slide 41 text
create 4core 8GB mem server !
Slide 42
Slide 42 text
benchmark.php
4147req/sec
fork_benchmark.php
3995req/sec
Slide 43
Slide 43 text
NO! why fork ver is more slow!? !
Slide 44
Slide 44 text
my httpd is BAD, that is why!!! !
I use nginx to benchmark my benchmark!! !
Slide 45
Slide 45 text
result(?) with nginx
4 benchmark.php
4 6817req/sec
4 fork_benchmark.php
4 13730req/sec
Slide 46
Slide 46 text
OK!
4 fork ver faster than not fork ver!!
Slide 47
Slide 47 text
( what is ok? )
Slide 48
Slide 48 text
meanwhile in apache bench...
4 ab -c 10 -n 10000 http://nginx/
4 Requests per second: 15273.21
Slide 49
Slide 49 text
compare own tool and ab.
4 fork_benchmark.php:
4 13,730 req/sec
4 ab:
4 15,273.21 #/sec
Slide 50
Slide 50 text
okey!(?)
Slide 51
Slide 51 text
conclusion
my bench (written by PHP) 0.9 times
faster than AB !!
( DISCLAIMER, THIS RESULT IS REAL RESULT. BUT JUST
PLAYING.)
Slide 52
Slide 52 text
OK !!
Slide 53
Slide 53 text
"what is ok?"
(now 2:12 am)
Slide 54
Slide 54 text
Other benchmarks.
4 php7.1.2 with my_httpd + fork_bench.php(c:10)
4 3858req/sec
4 php5.6.30 with my_httpd + fork_bench.php(c:10)
4 2425req/sec
4 hhvm3.18.1 with my_httpd + fork_bench.php(c:10)
4 2377req/sec
4 (but, hhvm w/o libevent)
Slide 55
Slide 55 text
conclusion
PHP7.1 is fast
(1.6x faster than 5.6 (at this test))
Slide 56
Slide 56 text
(REAL) conclusion
USE RELIABLE TOOLS
USE NGINX
DON'T USE PHP FOR WRITE A
BENCHMARK TOOL
Slide 57
Slide 57 text
but fun.
PHP is waiting New Challengers .....
FIN
Slide 58
Slide 58 text
//NEXT
use pthread;