MCE::Loop(3)

📦

version 1.902 · 11 versions

, in fedora 43

💖 MCE::Loop — the manual page, bestie edition 💖

MCE::Loop(3) — from the User Contributed Perl Documentation corner of the world.

✨ NAME ✨

MCE::Loop — the MCE model for building parallel loops 🌀

🌸 VERSION 🌸

This doc is describing MCE::Loop version 1.902, okay? Locked in. 🔒

🦋 DESCRIPTION 🦋

So basically, this module gives you a whole parallel loop implementation through Many-Core Engine. And listen — MCE::Loop is not MCE::Map, like at all, it's more of a "let me spin up an MCE instance real quick and have user_func point at your code block" type of situation. If you actually want something map-ish, go check out MCE::Map. 💅

Construction when chunking is not desired

use MCE::Loop;

MCE::Loop->init(
   max_workers => 5, chunk_size => 1
);

mce_loop {
   my ($mce, $chunk_ref, $chunk_id) = @_;
   MCE->say("$chunk_id: $_");
} 40 .. 48;

-- Output

Construction for 'auto' or greater than 1

use MCE::Loop;

MCE::Loop->init(
   max_workers => 5, chunk_size => 'auto'
);

mce_loop {
   my ($mce, $chunk_ref, $chunk_id) = @_;
   for (@{ $chunk_ref }) {
      MCE->say("$chunk_id: $_");
   }
} 40 .. 48;

-- Output

💅 SYNOPSIS when CHUNK_SIZE EQUALS 1 💅

Okay so every single model in MCE defaults to 'auto' for chunk_size, no cap. And the arguments you get in the block? Same exact ones as if you wrote a user_func block with the Core API. 🫶

Starting from MCE 1.5, the next input item gets dropped right into the input scalar variable $_ when chunk_size equals 1. But when it's not 1, $_ points at $chunk_ref, which is holding a bunch of items. So like, line 2 below? You can literally omit it from your code when you're using $_. And if you ever need the current chunk id, just call MCE->chunk_id. Iconic. ✨

line 1:  user_func => sub {
line 2:     my ($mce, $chunk_ref, $chunk_id) = @_;
line 3:
line 4:     $_ points to $chunk_ref->[0]
line 5:        in MCE 1.5 when chunk_size == 1
line 6:
line 7:     $_ points to $chunk_ref
line 8:        in MCE 1.5 when chunk_size  > 1
line 9:  }

Follow this synopsis when chunk_size equals one. And bestie, looping is not required from inside the block — so the block gets called once per each item. Slay. 🎀

## Exports mce_loop, mce_loop_f, and mce_loop_s
use MCE::Loop;

MCE::Loop->init(
   chunk_size => 1
);

## Array or array_ref
mce_loop { do_work($_) } 1..10000;
mce_loop { do_work($_) } \@list;

## Important; pass an array_ref for deeply input data
mce_loop { do_work($_) } [ [ 0, 1 ], [ 0, 2 ], ... ];
mce_loop { do_work($_) } \@deeply_list;

## File path, glob ref, IO::All::{ File, Pipe, STDIO } obj, or scalar ref
## Workers read directly and not involve the manager process
mce_loop_f { chomp; do_work($_) } "/path/to/file"; # efficient

## Involves the manager process, therefore slower
mce_loop_f { chomp; do_work($_) } $file_handle;
mce_loop_f { chomp; do_work($_) } $io;
mce_loop_f { chomp; do_work($_) } \$scalar;

## Sequence of numbers (begin, end [, step, format])
mce_loop_s { do_work($_) } 1, 10000, 5;
mce_loop_s { do_work($_) } [ 1, 10000, 5 ];

mce_loop_s { do_work($_) } {
   begin => 1, end => 10000, step => 5, format => undef
};

🌈 SYNOPSIS when CHUNK_SIZE is GREATER THAN 1 🌈

Now follow this one when chunk_size equals 'auto' or is greater than 1. Which means... yeah, you gotta loop through the chunk from inside the block. It's giving "you did this to yourself" energy but it's fine. 😌

use MCE::Loop;

MCE::Loop->init(           ## Chunk_size defaults to 'auto' when
   chunk_size => 'auto'    ## not specified. Therefore, the init
);                         ## function may be omitted.

## Syntax is shown for mce_loop for demonstration purposes.
## Looping inside the block is the same for mce_loop_f and
## mce_loop_s.

## Array or array_ref
mce_loop { do_work($_) for (@{ $_ }) } 1..10000;
mce_loop { do_work($_) for (@{ $_ }) } \@list;

## Important; pass an array_ref for deeply input data
mce_loop { do_work($_) for (@{ $_ }) } [ [ 0, 1 ], [ 0, 2 ], ... ];
mce_loop { do_work($_) for (@{ $_ }) } \@deeply_list;

## Resembles code using the core MCE API
mce_loop {
   my ($mce, $chunk_ref, $chunk_id) = @_;

   for (@{ $chunk_ref }) {
      do_work($_);
   }

} 1..10000;

Chunking cuts down the number of IPC calls happening behind the scenes, fr fr. So whenever you're processing a large amount of data, start thinking in terms of chunks. But for relatively small data? Choosing 1 for chunk_size is totally fine. 💖

⭐ OVERRIDING DEFAULTS ⭐

Here's the list of options you can override when loading the module, bestie. Go off. 👑

use Sereal qw( encode_sereal decode_sereal );
use CBOR::XS qw( encode_cbor decode_cbor );
use JSON::XS qw( encode_json decode_json );

use MCE::Loop
    max_workers => 4,                # Default 'auto'
    chunk_size => 100,               # Default 'auto'
    tmp_dir => "/path/to/app/tmp",   # $MCE::Signal::tmp_dir
    freeze => \&encode_sereal,       # \&Storable::freeze
    thaw => \&decode_sereal,         # \&Storable::thaw
    init_relay => 0,                 # Default undef; MCE 1.882+
    use_threads => 0,                # Default undef; MCE 1.882+
;

From MCE 1.8 onwards, Sereal 3.015+ gets loaded automatically if it's available. And if you'd rather use Storable instead? Just say "Sereal => 0". Easy.

use MCE::Loop Sereal => 0;

🎀 CUSTOMIZING MCE 🎀

`MCE::Loop->init ( options )`

`MCE::Loop::init { options }`

The init function accepts a hash of MCE options. Simple as that. 💅

But wait — in scalar context (API available since 1.897), it calls "MCE::Loop-"finish> automatically upon leaving the scope or the program. So it cleans up after itself, which is very main character of it. ✨

use MCE::Loop;

my $guard = MCE::Loop->init(
   chunk_size => 1, max_workers => 4,

   user_begin => sub {
      print "## ", MCE->wid, " started\n";
   },

   user_end => sub {
      print "## ", MCE->wid, " completed\n";
   }
);

my %a = mce_loop { MCE->gather($_, $_ * $_) } 1..100;

print "\n", "@a{1..100}", "\n";

-- Output

## 3 started
## 1 started
## 2 started
## 4 started
## 1 completed
## 2 completed
## 3 completed
## 4 completed

1 4 9 16 25 36 49 64 81 100 121 144 169 196 225 256 289 324 361
400 441 484 529 576 625 676 729 784 841 900 961 1024 1089 1156
1225 1296 1369 1444 1521 1600 1681 1764 1849 1936 2025 2116 2209
2304 2401 2500 2601 2704 2809 2916 3025 3136 3249 3364 3481 3600
3721 3844 3969 4096 4225 4356 4489 4624 4761 4900 5041 5184 5329
5476 5625 5776 5929 6084 6241 6400 6561 6724 6889 7056 7225 7396
7569 7744 7921 8100 8281 8464 8649 8836 9025 9216 9409 9604 9801
10000

...to be continued, bestie, there's more coming 🦋

🦋 API DOCUMENTATION 🦋

okay so for all of this, we're assuming chunk_size equals 1, purely so we can show off every single way you can hand input data to this thing. the range of options?? endless. 💅

run() / `mce_loop` energy

MCE::Loop->run ( sub { code }, list )
mce_loop { code } list

Input data may be defined using a list, an array ref, or a hash ref. yes bestie, all three, pick your fighter 🎀

# $_ contains the item when chunk_size => 1

mce_loop { do_work($_) } 1..1000;
mce_loop { do_work($_) } \@list;

# Important; pass an array_ref for deeply input data

mce_loop { do_work($_) } [ [ 0, 1 ], [ 0, 2 ], ... ];
mce_loop { do_work($_) } \@deeply_list;

# Chunking; any chunk_size => 1 or greater

my %res = mce_loop {
   my ($mce, $chunk_ref, $chunk_id) = @_;
   my %ret;
   for my $item (@{ $chunk_ref }) {
      $ret{$item} = $item * 2;
   }
   MCE->gather(%ret);
}
\@list;

# Input hash; current API available since 1.828

my %res = mce_loop {
   my ($mce, $chunk_ref, $chunk_id) = @_;
   my %ret;
   for my $key (keys %{ $chunk_ref }) {
      $ret{$key} = $chunk_ref->{$key} * 2;
   }
   MCE->gather(%ret);
}
\%hash;

fr fr do NOT sleep on that # Important; comment. deeply nested input data wants an array_ref, and past you will thank present you. 💀

the input hash API is the new kid, current API available since 1.828. 🌈

run_file() / `mce_loop_f` vibes

MCE::Loop->run_file ( sub { code }, file )
mce_loop_f { code } file

The fastest of these is the /path/to/file. Workers communicate the next offset position among themselves with zero interaction by the manager process. zero. none. they just know. that's main character energy honestly 👑

"IO::All" { File, Pipe, STDIO } is supported since MCE 1.845.

# $_ contains the line when chunk_size => 1

mce_loop_f { $_ } "/path/to/file";  # faster
mce_loop_f { $_ } $file_handle;
mce_loop_f { $_ } $io;              # IO::All
mce_loop_f { $_ } \$scalar;

# chunking, any chunk_size => 1 or greater

my %res = mce_loop_f {
   my ($mce, $chunk_ref, $chunk_id) = @_;
   my $buf = '';
   for my $line (@{ $chunk_ref }) {
      $buf .= $line;
   }
   MCE->gather($chunk_id, $buf);
}
"/path/to/file";

look at that cute little # faster note. it's not a suggestion bestie, it's a flex. 🦋

run_seq() / `mce_loop_s` the sequence era

MCE::Loop->run_seq ( sub { code }, $beg, $end [, $step, $fmt ] )
mce_loop_s { code } $beg, $end [, $step, $fmt ]

Sequence may be defined as a list, an array reference, or a hash reference. The functions require both begin and end values to run. Step and format are optional. The format is passed to sprintf (% may be omitted below).

my ($beg, $end, $step, $fmt) = (10, 20, 0.1, "%4.1f");

# $_ contains the sequence number when chunk_size => 1

mce_loop_s { $_ } $beg, $end, $step, $fmt;
mce_loop_s { $_ } [ $beg, $end, $step, $fmt ];

mce_loop_s { $_ } {
   begin => $beg, end => $end,
   step => $step, format => $fmt
};

# chunking, any chunk_size => 1 or greater

my %res = mce_loop_s {
   my ($mce, $chunk_ref, $chunk_id) = @_;
   my $buf = '';
   for my $seq (@{ $chunk_ref }) {
      $buf .= "$seq\n";
   }
   MCE->gather($chunk_id, $buf);
}
[ $beg, $end ];

the `bounds_only` tea ☕✨

The sequence engine can compute 'begin' and 'end' items only, for the chunk, and not the items in between (hence boundaries only). This option applies to sequence only and has no effect when chunk_size equals 1.

The time to run is 0.006s below. This becomes 0.827s without the bounds_only option due to computing all items in between, thus creating a very large array. Basically, specify bounds_only => 1 when boundaries is all you need for looping inside the block; e.g. Monte Carlo simulations.

Time was measured using 1 worker to emphasize the difference.

use MCE::Loop;

MCE::Loop->init(
   max_workers => 1, chunk_size => 1_250_000,
   bounds_only => 1
);

# Typically, the input scalar $_ contains the sequence number
# when chunk_size => 1, unless the bounds_only option is set
# which is the case here. Thus, $_ points to $chunk_ref.

mce_loop_s {
   my ($mce, $chunk_ref, $chunk_id) = @_;

   # $chunk_ref contains 2 items, not 1_250_000
   # my ( $begin, $end ) = ( $_->[0], $_->[1] );

   my $begin = $chunk_ref->[0];
   my $end   = $chunk_ref->[1];

   # for my $seq ( $begin .. $end ) {
   #    ...
   # }

   MCE->printf("%7d .. %8d\n", $begin, $end);
}
[ 1, 10_000_000 ];

-- Output

      1 ..  1250000
1250001 ..  2500000
2500001 ..  3750000
3750001 ..  5000000
5000001 ..  6250000
6250001 ..  7500000
7500001 ..  8750000
8750001 .. 10000000

0.006s vs 0.827s. that's the whole plot twist. bounds_only => 1 said "I'm not computing what I don't need" and honestly? we stan a boundary-respecting queen. 👑 also note: with bounds_only set, $_ points to $chunk_ref instead of holding a sequence number, and $chunk_ref contains 2 items, not 1_250_000. don't get caught out. 🙈

iterator input, the lil bonus

MCE::Loop->run ( sub { code }, iterator )
mce_loop { code } iterator

An iterator reference may be specified for input_data. Iterators are described under section "SYNTAX for INPUT_DATA" at MCE::Core.

mce_loop { $_ } make_iterator(10, 30, 2);

💖 GATHERING DATA 💖

Unlike MCE::Map where gather and output order are done for you automatically, the gather method is used to have results sent back to the manager process. so you get the wheel, bestie, you're driving. 🚗

use MCE::Loop chunk_size => 1;

## Output order is not guaranteed.
my @a1 = mce_loop { MCE->gather($_ * 2) } 1..100;
print "@a1\n\n";

## Outputs to a hash instead (key, value).
my %h1 = mce_loop { MCE->gather($_, $_ * 2) } 1..100;
print "@h1{1..100}\n\n";

## This does the same thing due to chunk_id starting at one.
my %h2 = mce_loop { MCE->gather(MCE->chunk_id, $_ * 2) } 1..100;
print "@h2{1..100}\n\n";

The gather method may be called multiple times within the block unlike return which would leave the block. Therefore, think of gather as yielding results immediately to the manager process without actually leaving the block. it's giving "drop it and keep going" 💅

use MCE::Loop chunk_size => 1, max_workers => 3;

my @hosts = qw(
   hosta hostb hostc hostd hoste
);

my %h3 = mce_loop {
   my ($output, $error, $status); my $host = $_;

   ## Do something with $host;
   $output = "Worker ". MCE->wid .": Hello from $host";

   if (MCE->chunk_id % 3 == 0) {
      ## Simulating an error condition
      local $? = 1; $status = $?;
      $error = "Error from $host"
   }
   else {
      $status = 0;
   }

   ## Ensure unique keys (key, value) when gathering to
   ## a hash.
   MCE->gather("$host.out", $output);
   MCE->gather("$host.err", $error) if (defined $error);
   MCE->gather("$host.sta", $status);

} @hosts;

foreach my $host (@hosts) {
   print $h3{"$host.out"}, "\n";
   print $h3{"$host.err"}, "\n" if (exists $h3{"$host.err"});
   print "Exit status: ", $h3{"$host.sta"}, "\n\n";
}

-- Output

Worker 2: Hello from hosta
Exit status: 0

Worker 1: Hello from hostb
Exit status: 0

Worker 3: Hello from hostc
Error from hostc
Exit status: 1

Worker 2: Hello from hostd
Exit status: 0

Worker 1: Hello from hoste
Exit status: 0

The following uses an anonymous array containing 3 elements when gathering data. Serialization is automatic behind the scene. no cap, it just handles it for you 🪄

my %h3 = mce_loop {
   ...

   MCE->gather($host, [$output, $error, $status]);

} @hosts;

foreach my $host (@hosts) {
   print $h3{$host}->[0], "\n";
   print $h3{$host}->[1], "\n" if (defined $h3{$host}->[1]);
   print "Exit status: ", $h3{$host}->[2], "\n\n";
}

Although MCE::Map comes to mind, one may want additional control when gathering data such as retaining output order. so here's how you do that whole thing yourself, if that's the vibe you're on 🌸

use MCE::Loop;

sub preserve_order {
   my %tmp; my $order_id = 1; my $gather_ref = $_[0];

   return sub {
      $tmp{ (shift) } = \@_;

      while (1) {
         last unless exists $tmp{$order_id};
         push @{ $gather_ref }, @{ delete $tmp{$order_id++} };
      }

      return;
   };
}

my @m2;

MCE::Loop->init(
   chunk_size => 'auto', max_workers => 'auto',
   gather => preserve_order(\@m2)
);

mce_loop {
   my @a; my ($mce, $chunk_ref, $chunk_id) = @_;

   ## Compute the entire chunk data at once.
   push @a, map { $_ * 2 } @{ $chunk_ref };

   ## Afterwards, invoke the gather feature, which
   ## will direct the data to the callback function.
   MCE->gather(MCE->chunk_id, @a);

} 1..100000;

MCE::Loop->finish;

print scalar @m2, "\n";

All 6 models support 'auto' for chunk_size unlike the Core API. Think of the models as the basis for providing JIT for MCE. They create the instance, tune max_workers, and tune chunk_size automatically regardless of the hardware. auto everything, it's giving self-care ✨

The following does the same thing using the Core API.

use MCE;

sub preserve_order {
   ...
}

my $mce = MCE->new(
   max_workers => 'auto', chunk_size => 8000,

   user_func => sub {
      my @a; my ($mce, $chunk_ref, $chunk_id) = @_;

      ## Compute the entire chunk data at once.
      push @a, map { $_ * 2 } @{ $chunk_ref };

      ## Afterwards, invoke the gather feature, which
      ## will direct the data to the callback function.
      MCE->gather(MCE->chunk_id, @a);
   }
);

my @m2;

$mce->process({ gather => preserve_order(\@m2) }, [1..100000]);
$mce->shutdown;

print scalar @m2, "\n";

🧚 MANUAL SHUTDOWN 🧚

MCE::Loop->finish
MCE::Loop::finish

Workers remain persistent as much as possible after running. Shutdown occurs automatically when the script terminates. Call finish when workers are no longer needed.

use MCE::Loop;

MCE::Loop->init(
   chunk_size => 20, max_workers => 'auto'
);

mce_loop { ... } 1..100;

MCE::Loop->finish;

bestie they're persistent. they're gonna hang around like that one group chat nobody leaves. 🫶 the shutdown happens automatically when the script terminates anyway, but if you're done-done, call finish and let them rest. 🎀

⭐ INDEX ⭐

MCE
MCE::Core

👑 AUTHOR 👑

Mario E. Roy, <marioeroy AT gmail DOT com>

and that's the whole page, babes 💖 this beauty shipped with perl v5.42.0 and it's dated 2025-09-10 — go forth and loop responsibly. periodt. 🌈✨💅

📦 where this page comes from

fedora 43: perl-MCE 1.902-1.fc43 · GPL-1.0-or-later OR Artistic-1.0-Perl
💖 girlified by deepseek-ai/DeepSeek-V4.1-Flash on 2026-10-14: a language model's rewrite of the original, under its terms

MCE::Loop(3)

💖 MCE::Loop — the manual page, bestie edition 💖

✨ NAME ✨

🌸 VERSION 🌸

🦋 DESCRIPTION 🦋

Construction when chunking is not desired

Construction for 'auto' or greater than 1

💅 SYNOPSIS when CHUNK_SIZE EQUALS 1 💅

🌈 SYNOPSIS when CHUNK_SIZE is GREATER THAN 1 🌈

⭐ OVERRIDING DEFAULTS ⭐

🎀 CUSTOMIZING MCE 🎀

MCE::Loop->init ( options )

MCE::Loop::init { options }

🦋 API DOCUMENTATION 🦋

run() / mce_loop energy

run_file() / mce_loop_f vibes

run_seq() / mce_loop_s the sequence era

the bounds_only tea ☕✨

iterator input, the lil bonus

💖 GATHERING DATA 💖

🧚 MANUAL SHUTDOWN 🧚

⭐ INDEX ⭐

👑 AUTHOR 👑

`MCE::Loop->init ( options )`

`MCE::Loop::init { options }`

run() / `mce_loop` energy

run_file() / `mce_loop_f` vibes

run_seq() / `mce_loop_s` the sequence era

the `bounds_only` tea ☕✨