1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # ParallelInexport.pm -- useful class to support parallel_import.pl
|
---|
4 | # A component of the Greenstone digital library software
|
---|
5 | # from the New Zealand Digital Library Project at the
|
---|
6 | # University of Waikato, New Zealand.
|
---|
7 | #
|
---|
8 | # Copyright (C) 1999 New Zealand Digital Library Project
|
---|
9 | #
|
---|
10 | # This program is free software; you can redistribute it and/or modify
|
---|
11 | # it under the terms of the GNU General Public License as published by
|
---|
12 | # the Free Software Foundation; either version 2 of the License, or
|
---|
13 | # (at your option) any later version.
|
---|
14 | #
|
---|
15 | # This program is distributed in the hope that it will be useful,
|
---|
16 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
17 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
18 | # GNU General Public License for more details.
|
---|
19 | #
|
---|
20 | # You should have received a copy of the GNU General Public License
|
---|
21 | # along with this program; if not, write to the Free Software
|
---|
22 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
23 | #
|
---|
24 | ###########################################################################
|
---|
25 |
|
---|
26 | package ParallelInexport;
|
---|
27 |
|
---|
28 | use strict;
|
---|
29 |
|
---|
30 | # Randomize the order of files in the filelist
|
---|
31 | use List::Util qw( shuffle );
|
---|
32 |
|
---|
33 | # index the files in parallel using MPI farmer to farm off multiple processes
|
---|
34 | # [hs, 1 july 2010]
|
---|
35 | sub farm_out_processes
|
---|
36 | {
|
---|
37 | my ($jobs, $epoch, $importdir, $block_hash, $collection, $site) = @_;
|
---|
38 |
|
---|
39 | my $tmp_dir_path = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, 'tmp');
|
---|
40 | if (!-d $tmp_dir_path)
|
---|
41 | {
|
---|
42 | mkdir($tmp_dir_path, 0777);
|
---|
43 | }
|
---|
44 |
|
---|
45 | # create the list of files to import
|
---|
46 | my $overwrite = 1;
|
---|
47 | my $tmp_filelist = &util::filename_cat($tmp_dir_path, "filelist.txt");
|
---|
48 | # - if the file is already there (which is should be during testing) then
|
---|
49 | # don't regenerate. This is especially important for imports of 1 million
|
---|
50 | # documents as just the directory scan can take several hours.
|
---|
51 | if ($overwrite || !-f $tmp_filelist)
|
---|
52 | {
|
---|
53 | open (my $filelist, ">$tmp_filelist");
|
---|
54 | my @filenames = keys %{$block_hash->{'all_files'}};
|
---|
55 | @filenames = shuffle(@filenames);
|
---|
56 | foreach my $filename (@filenames)
|
---|
57 | {
|
---|
58 | my $full_filename = &util::filename_cat($importdir,$filename);
|
---|
59 | if ((! exists $block_hash->{'file_blocks'}->{$full_filename})
|
---|
60 | && ($filename !~ m/metadata\.xml$/))
|
---|
61 | {
|
---|
62 | print $filelist "$filename\n";
|
---|
63 | }
|
---|
64 | }
|
---|
65 | close ($filelist);
|
---|
66 | }
|
---|
67 |
|
---|
68 | # Determine if we've been provided a mpi.conf file to indicate the other
|
---|
69 | # machines (slave nodes) this parallizable process should run on
|
---|
70 | my $mpi_conf_path = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, 'mpi.conf');
|
---|
71 | my $mpi_flags = '';
|
---|
72 | if (-f $mpi_conf_path)
|
---|
73 | {
|
---|
74 | print STDERR " ***** CLUSTER MODE *****\n";
|
---|
75 | $mpi_flags .= '-machinefile "' . $mpi_conf_path . '" ';
|
---|
76 | #$mpi_flags .= '-nolocal ';
|
---|
77 | }
|
---|
78 | else
|
---|
79 | {
|
---|
80 | print STDERR " ***** SINGLE COMPUTER MODE *****\n";
|
---|
81 | }
|
---|
82 | $mpi_flags .= ' --show-progress --timestamp-output --verbose';
|
---|
83 |
|
---|
84 | # invoke the farmer to start processing the files
|
---|
85 | $site = "" if (!defined $site);
|
---|
86 | my $gsdlhome = $ENV{'GSDLHOME'};
|
---|
87 | my $farmer_exe = $gsdlhome . '/ext/parallel-building/linux/bin/mpiimport';
|
---|
88 | my $mpi_cmd = $gsdlhome . '/ext/parallel-building/linux/bin/mpirun ' . $mpi_flags . ' -n ' . $jobs . ' ' . $farmer_exe . ' ' . $tmp_filelist . ' ' . $epoch . ' ' . $gsdlhome . ' ' . $collection . ' ' . $site;
|
---|
89 | print STDERR "MPI Command: \"" . $mpi_cmd . "\"\n";
|
---|
90 | # system ($mpi_cmd);
|
---|
91 | open(MPI, $mpi_cmd . " |") or die("Couldn't Execute MPI");
|
---|
92 | while ( defined( my $line = <MPI> ) )
|
---|
93 | {
|
---|
94 | chomp($line);
|
---|
95 | print "$line\n";
|
---|
96 | }
|
---|
97 | close(MPI);
|
---|
98 | }
|
---|
99 |
|
---|
100 | 1;
|
---|