Pipeline of Redis Performance improvement

1, the normal use of the process before

The client sends a query to the server and reads from the socket, usually in a blocking manner, for the server response.
The server processes the command and sends the response back to the client.
Which means that every command will have a previous process.

2, the significance of the pipeline
If the successive Redis commands can be returned uniformly after the operation has been completed, the number of connections can be reduced, and the delay time will be reduced, and then the pipeline will be generated.
The basic implication of the pipeline is that the client can send multiple requests to the server without waiting for a reply and eventually reading the reply in one step.

3. Parameter Description:
Redis::multi or Redis::P ipeline. Default is Redis::multi
Redis::multi: Perform multiple operations as one transaction
Redis::P ipeline: Let (multiple) execute commands in a simple, faster way to send to the server, but without any guarantee of atomicity

Test Code 1:

<?phpset_time_limit (0); Ini_set (' Memory_limit ', ' 1024M '); $redis = new Redis (); G (' 1 '); $redis->connect (' 127.0.0.1 ');//not Atomic, pipe $redis->pipeline (); for ($i = 0; $i < 100000; $i + +) {$redis-&G    T;set ("test_{$i}", Pow ($i, 2)); $redis->get ("test_{$i}");} $redis->exec (); $redis->close (); G (' 1 ', ' e '); G (' 2 '); $redis->connect (' 127.0.0.1 ');//Things have atomic $redis->multi (); for ($i = 0; $i < 100000; $i + +) {$redis->set (    "Test_{$i}", Pow ($i, 2)); $redis->get ("test_{$i}");} $redis->exec (); $redis->close (); G (' 2 ', ' e ');//General G (' 3 '); $redis->connect (' 127.0.0.1 ');//Things have atomic for ($i = 0; $i < 100000; $i + +) {$redis->set ("Te    st_{$i} ", Pow ($i, 2)); $redis->get ("test_{$i}");} $redis->close ();    G (' 3 ', ' e '); function G ($star, $end = ") {Static $info = array ();        if (!empty ($end)) {$info [$end] = Microtime (true);        $sconds = $info [$end]-$info [$star];        Echo $sconds, "MS";    Echo Php_eol; } else {$info [$star] = Microtime (true); }}

Result output:
0.85941696166992ms 14.938266992569ms 15.121881961823ms

Test Code 2:

<?php$redis = new Redis();$redis->connect(‘127.0.0.1‘, 6379);$pipe = $redis->multi(Redis::PIPELINE);for ($i = 0; $i < 3; $i++) {    $key = "key::{$i}";    print_r($pipe->set($key, str_pad($i, 2, ‘0‘, 0)));    echo PHP_EOL;    print_r($pipe->get($key));    echo PHP_EOL;}$result = $pipe->exec();echo "<pre>";var_dump($result);

Output Result:
Redis object () Redis object () Redis object () Redis object () Redis object () Redis object ()
Array (6) {
[0]=>
BOOL (TRUE)
[1]=>
String (2) "00"
[2]=>
BOOL (TRUE)
[3]=>
String (2) "01"
[4]=>
BOOL (TRUE)
[5]=>
String (2) "02"
}

Pipeline of Redis Performance improvement