view release on metacpan or  search on metacpan

README.md  view on Meta::CPAN

[![Actions Status](https://github.com/plainbanana/Redis-Cluster-Fast/workflows/test/badge.svg)](https://github.com/plainbanana/Redis-Cluster-Fast/actions) [![MetaCPAN Release](https://badge.fury.io/pl/Redis-Cluster-Fast.svg)](https://metacpan.org/rel...
# NAME

Redis::Cluster::Fast - A fast perl binding for Redis Cluster

# SYNOPSIS

    use Redis::Cluster::Fast;

    Redis::Cluster::Fast::srandom(100);

    my $redis = Redis::Cluster::Fast->new(
        startup_nodes => [
            'localhost:9000',
            'localhost:9001',
            'localhost:9002',
            'localhost:9003',
            'localhost:9004',
            'localhost:9005',
        ],
        connect_timeout => 0.05,
        command_timeout => 0.05,
        max_retry_count => 10,
    );

    $redis->set('test', 123);

    # '123'
    my $str = $redis->get('test');

    $redis->mset('{my}foo', 'hoge', '{my}bar', 'fuga');

    # get as array-ref
    my $array_ref = $redis->mget('{my}foo', '{my}bar');
    # get as array
    my @array = $redis->mget('{my}foo', '{my}bar');

    $redis->hset('mymap', 'field1', 'Hello');
    $redis->hset('mymap', 'field2', 'ByeBye');

    # get as hash-ref
    my $hash_ref = { $redis->hgetall('mymap') };
    # get as hash
    my %hash = $redis->hgetall('mymap');

# DESCRIPTION

Redis::Cluster::Fast is like [Redis::Fast](https://github.com/shogo82148/Redis-Fast) but support Redis Cluster by [hiredis-cluster](https://github.com/Nordix/hiredis-cluster).

To build and use this module you need libevent-dev >= 2.x is installed on your system.

Recommend Redis 6 or higher.

Since Redis 6, it supports new version of Redis serialization protocol, [RESP3](https://github.com/antirez/RESP3/blob/master/spec.md).
This client start to connect using RESP2 and currently it has no option to upgrade all connections to RESP3.

## MICROBENCHMARK

Simple microbenchmark comparing PP and XS.
The benchmark script used can be found under examples directory.
Each operation was executed 100,000 times, and the execution time was measured in milliseconds.

    +--------------------------------+-------+-------+-------+-------+-------+
    | Operation                      | P50   | P80   | P95   | P99   | P100  |
    +--------------------------------+-------+-------+-------+-------+-------+
    | get_pp                         | 0.028 | 0.032 | 0.036 | 0.050 | 0.880 |
    | get_xs                         | 0.020 | 0.023 | 0.025 | 0.044 | 0.881 |
    | get_xs_pipeline                | 0.014 | 0.015 | 0.018 | 0.021 | 0.472 |
    | get_xs_pipeline_batched_100    | 0.003 | 0.003 | 0.004 | 0.074 | 0.323 |
    | set_pp                         | 0.028 | 0.032 | 0.037 | 0.051 | 2.014 |
    | set_xs                         | 0.021 | 0.024 | 0.027 | 0.047 | 0.729 |
    | set_xs_pipeline                | 0.014 | 0.016 | 0.018 | 0.021 | 0.393 |
    | set_xs_pipeline_batched_100    | 0.003 | 0.004 | 0.005 | 0.073 | 0.379 |
    +--------------------------------+-------+-------+-------+-------+-------+

c.f. https://github.com/plainbanana/Redis-Cluster-Fast-Benchmarks

# METHODS

## srandom($seed)

hiredis-cluster uses [random()](https://linux.die.net/man/3/random) to select a node used for requesting cluster topology.