Redis::Fast(3pm)

NAME

Redis::Fast - Perl binding for Redis database

SYNOPSIS

    ## Defaults to $ENV{REDIS_SERVER} or 127.0.0.1:6379
    my $redis = Redis::Fast->new;
    my $redis = Redis::Fast->new(server => 'redis.example.com:8080');
    ## Set the connection name (requires Redis 2.6.9)
    my $redis = Redis::Fast->new(
      server => 'redis.example.com:8080',
      name => 'my_connection_name',
    );
    my $generation = 0;
    my $redis = Redis::Fast->new(
      server => 'redis.example.com:8080',
      name => sub { "cache-$$-".++$generation },
    );
    ## Use Sentinels, possibly with password
    my $redis = Redis::Fast->new(
      sentinels => [ '10.0.0.1:16379', '10.0.0.2:16379', ],
      service   => 'mymaster',
      sentinels_password => 'TheB1gS3CR3T', # optional
    );
    ## Use UNIX domain socket
    my $redis = Redis::Fast->new(sock => '/path/to/socket');
    ## Enable auto-reconnect
    ## Try to reconnect every 500ms up to 60 seconds until success
    ## Die if you can't after that
    my $redis = Redis::Fast->new(reconnect => 60, every => 500_000);
    ## Try each 100ms up to 2 seconds (every is in microseconds)
    my $redis = Redis::Fast->new(reconnect => 2, every => 100_000);
    ## Disable the automatic utf8 encoding => much more performance
    ## !!!! This will be the default after 2.000, see ENCODING below
    my $redis = Redis::Fast->new(encoding => undef);
    ## Use all the regular Redis commands, they all accept a list of
    ## arguments
    ## See http://redis.io/commands for full list
    $redis->get('key');
    $redis->set('key' => 'value');
    $redis->sort('list', 'DESC');
    $redis->sort(qw{list LIMIT 0 5 ALPHA DESC});
    ## Add a coderef argument to run a command in the background
    $redis->sort(qw{list LIMIT 0 5 ALPHA DESC}, sub {
      my ($reply, $error) = @_;
      die "Oops, got an error: $error\n" if defined $error;
      print "$_\n" for @$reply;
    });
    long_computation();
    $redis->wait_all_responses;
    ## or
    $redis->wait_one_response();
    ## Or run a large batch of commands in a pipeline
    my %hash = _get_large_batch_of_commands();
    $redis->hset('h', $_, $hash{$_}, sub {}) for keys %hash;
    $redis->wait_all_responses;
    ## Publish/Subscribe
    $redis->subscribe(
      'topic_1',
      'topic_2',
      sub {
        my ($message, $topic, $subscribed_topic) = @_
          ## $subscribed_topic can be different from topic if
          ## you use psubscribe() with wildcards
      }
    );
    $redis->psubscribe('nasdaq.*', sub {...});
    ## Blocks and waits for messages, calls subscribe() callbacks
    ##  ... forever
    my $timeout = 10;
    $redis->wait_for_messages($timeout) while 1;
    ##  ... until some condition
    my $keep_going = 1; ## other code will set to false to quit
    $redis->wait_for_messages($timeout) while $keep_going;
    $redis->publish('topic_1', 'message');

DESCRIPTION

"Redis::Fast" is a wrapper around Salvatore Sanfilippo's hiredis <https://github.com/redis/hiredis> C client. It is compatible with Redis.pm <https://github.com/melo/perl-redis>.

This version supports protocol 2.x (multi-bulk) or later of Redis available at <https://github.com/antirez/redis/>.

Reconnect on error

Besides auto-reconnect when the connection is closed, "Redis::Fast" supports reconnecting on the specified errors by the "reconnect_on_error" option. Here's an example that will reconnect when receiving "READONLY" error:

    my $r = Redis::Fast->new(
      reconnect          => 1, # The value greater than 0 is required
      reconnect_on_error => sub {
        my ($error, $ret, $command) = @_;
        if ($error =~ /READONLY You can't write against a read only slave/) {
          # force reconnect
          return 1;
        }
        # do nothing
        return -1;
      },
    );

This feature is useful when using Amazon ElastiCache. Once failover happens, Amazon ElastiCache will switch the master we currently connected with to a slave, leading to the following writes fails with the error "READONLY". Using "reconnect_on_error", we can force the connection to reconnect on this error in order to connect to the new master. If your Elasticache Redis is enabled to be set an option for close-on-slave-write <https://docs.aws.amazon.com/AmazonElastiCache/latest/UserGuide/ParameterGroups.Redis.html#ParameterGroups.Redis.2-8-23>, this feature might be unnecessary.

The return value of "reconnect_on_error" should be greater than -2. -1 means that "Redis::Fast" behaves the same as without this option. 0 and greater than 0 means that "Redis::Fast" forces to reconnect and then wait for a next force reconnect until this value seconds elapse. This unit is a second, and the type is double. For example, 0.01 means 10 milliseconds.

Note: This feature is not supported for the subscribed mode.

PERFORMANCE IN SYNCHRONIZE MODE

Redis.pm

    Benchmark: running 00_ping, 10_set, 11_set_r, 20_get, 21_get_r, 30_incr, 30_incr_r, 40_lpush, 50_lpop, 90_h_get, 90_h_set for at least 5 CPU seconds...
       00_ping:  8 wallclock secs ( 0.69 usr +  4.77 sys =  5.46 CPU) @ 5538.64/s (n=30241)
        10_set:  8 wallclock secs ( 1.07 usr +  4.01 sys =  5.08 CPU) @ 5794.09/s (n=29434)
      11_set_r:  7 wallclock secs ( 0.42 usr +  4.84 sys =  5.26 CPU) @ 5051.33/s (n=26570)
        20_get:  8 wallclock secs ( 0.69 usr +  4.82 sys =  5.51 CPU) @ 5080.40/s (n=27993)
      21_get_r:  7 wallclock secs ( 2.21 usr +  3.09 sys =  5.30 CPU) @ 5389.06/s (n=28562)
       30_incr:  7 wallclock secs ( 0.69 usr +  4.73 sys =  5.42 CPU) @ 5671.77/s (n=30741)
     30_incr_r:  7 wallclock secs ( 0.85 usr +  4.31 sys =  5.16 CPU) @ 5824.42/s (n=30054)
      40_lpush:  8 wallclock secs ( 0.60 usr +  4.77 sys =  5.37 CPU) @ 5832.59/s (n=31321)
       50_lpop:  7 wallclock secs ( 1.24 usr +  4.17 sys =  5.41 CPU) @ 5112.75/s (n=27660)
      90_h_get:  7 wallclock secs ( 0.63 usr +  4.65 sys =  5.28 CPU) @ 5716.29/s (n=30182)
      90_h_set:  7 wallclock secs ( 0.65 usr +  4.74 sys =  5.39 CPU) @ 5593.14/s (n=30147)

Redis::Fast

Redis::Fast is 50% faster than Redis.pm.

    Benchmark: running 00_ping, 10_set, 11_set_r, 20_get, 21_get_r, 30_incr, 30_incr_r, 40_lpush, 50_lpop, 90_h_get, 90_h_set for at least 5 CPU seconds...
       00_ping:  9 wallclock secs ( 0.18 usr +  4.84 sys =  5.02 CPU) @ 7939.24/s (n=39855)
        10_set: 10 wallclock secs ( 0.31 usr +  5.40 sys =  5.71 CPU) @ 7454.64/s (n=42566)
      11_set_r:  9 wallclock secs ( 0.31 usr +  4.87 sys =  5.18 CPU) @ 7993.05/s (n=41404)
        20_get: 10 wallclock secs ( 0.27 usr +  4.84 sys =  5.11 CPU) @ 8350.68/s (n=42672)
      21_get_r: 10 wallclock secs ( 0.32 usr +  5.17 sys =  5.49 CPU) @ 8238.62/s (n=45230)
       30_incr:  9 wallclock secs ( 0.23 usr +  5.27 sys =  5.50 CPU) @ 8221.82/s (n=45220)
     30_incr_r:  8 wallclock secs ( 0.28 usr +  4.91 sys =  5.19 CPU) @ 8092.29/s (n=41999)
      40_lpush:  9 wallclock secs ( 0.18 usr +  5.06 sys =  5.24 CPU) @ 8312.02/s (n=43555)
       50_lpop:  9 wallclock secs ( 0.20 usr +  4.84 sys =  5.04 CPU) @ 8010.12/s (n=40371)
      90_h_get:  9 wallclock secs ( 0.19 usr +  5.51 sys =  5.70 CPU) @ 7467.72/s (n=42566)
      90_h_set:  8 wallclock secs ( 0.28 usr +  4.83 sys =  5.11 CPU) @ 7724.07/s (n=39470)o

PERFORMANCE IN PIPELINE MODE

    #!/usr/bin/perl
    use warnings;
    use strict;
    use Time::HiRes qw/time/;
    use Redis;
    my $count = 100000;
    {
        my $r = Redis->new;
        my $start = time;
        for(1..$count) {
            $r->set('hoge', 'fuga', sub{});
        }
        $r->wait_all_responses;
        printf "Redis.pm:\n%.2f/s\n", $count / (time - $start);
    }
    {
        my $r = Redis::Fast->new;
        my $start = time;
        for(1..$count) {
            $r->set('hoge', 'fuga', sub{});
        }
        $r->wait_all_responses;
        printf "Redis::Fast:\n%.2f/s\n", $count / (time - $start);
    }

Redis::Fast is 4x faster than Redis.pm in pipeline mode.

    Redis.pm:
    22588.95/s
    Redis::Fast:
    81098.01/s

AUTHOR

Ichinose Shogo <[email protected]>

LICENSE

This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself.