Queen.pm 13.7 KB
Newer Older
Jessica Severin's avatar
Jessica Severin committed
1
2
3
4
5
6
7
8
#
# You may distribute this module under the same terms as perl itself
#
# POD documentation - main docs before the code

=pod 

=head1 NAME
9
  Bio::EnsEMBL::Hive::Queen
Jessica Severin's avatar
Jessica Severin committed
10
11

=head1 SYNOPSIS
12
  The Queen of the Hive based job control system
Jessica Severin's avatar
Jessica Severin committed
13
14

=head1 DESCRIPTION
15
16
17
18
19
20
21
22
23
24
25
26
  The Queen of the Hive based job control system is responsible to 'birthing' the
  correct number of workers of the right type so that they can find jobs to do.
  It will also free up jobs of Workers that died unexpectantly so that other workers
  can claim them to do.

  Hive based processing is a concept based on a more controlled version
  of an autonomous agent type system.  Each worker is not told what to do
  (like a centralized control system - like the current pipeline system)
  but rather queries a central database for jobs (give me jobs).

  Each worker is linked to an analysis_id, registers its self on creation
  into the Hive, creates a RunnableDB instance of the Analysis->module,
27
  gets $worker->batch_size() jobs from the analysis_job table, does its
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
  work, creates the next layer of analysis_job entries by interfacing to
  the DataflowRuleAdaptor to determine the analyses it needs to pass it's
  output data to and creates jobs on the next analysis's database.
  It repeats this cycle until it's lived it's lifetime or until there are no
  more jobs left.
  The lifetime limit is just a safety limit to prevent these from 'infecting'
  a system.

  The Queens job is to simply birth Workers of the correct analysis_id to get the
  work down.  The only other thing the Queen does is free up jobs that were
  claimed by Workers that died unexpectantly so that other workers can take
  over the work.

  The Beekeeper is in charge of interfacing between the Queen and a compute resource
  or 'compute farm'.  It's job is to query Queens if they need any workers and to
  send the requested number of workers to open machines via the runWorker.pl script.
  It is also responsible for interfacing with the Queen to identify worker which died
  unexpectantly.
Jessica Severin's avatar
Jessica Severin committed
46
47

=head1 CONTACT
48
49
  Contact Jessica Severin on EnsEMBL::Hive implemetation/design detail: jessica@ebi.ac.uk
  Contact Ewan Birney on EnsEMBL in general: birney@sanger.ac.uk
Jessica Severin's avatar
Jessica Severin committed
50
51

=head1 APPENDIX
52
53
  The rest of the documentation details each of the object methods. 
  Internal methods are usually preceded with a _
Jessica Severin's avatar
Jessica Severin committed
54
55
56
57
58
59
60
61
62
63

=cut

package Bio::EnsEMBL::Hive::Queen;

use strict;
use strict;
use Bio::EnsEMBL::Hive::Worker;
use Bio::EnsEMBL::DBSQL::BaseAdaptor;
use Sys::Hostname;
64
use Bio::EnsEMBL::Hive::DBSQL::AnalysisCtrlRuleAdaptor;
Jessica Severin's avatar
Jessica Severin committed
65
66
67
68

our @ISA = qw(Bio::EnsEMBL::DBSQL::BaseAdaptor);

#
69
# PUBLIC METHODS
Jessica Severin's avatar
Jessica Severin committed
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
#
################

=head2 create_new_worker

  Arg [1]    : $analysis_id
  Example    :
  Description:
  Returntype : Bio::EnsEMBL::Hive::Worker
  Exceptions :
  Caller     :

=cut

sub create_new_worker {
85
86
87
  my $self        = shift;
  my $analysis_id = shift;
  my $beekeeper   = shift;
Jessica Severin's avatar
Jessica Severin committed
88

89
  my $analStatsDBA = $self->db->get_AnalysisStatsAdaptor;
90
91
  return undef unless($analStatsDBA);
  
92
  my $analysisStats = $analStatsDBA->fetch_by_analysis_id($analysis_id);
93
94
95
96
97
98
99
100
101
102
  return undef unless($analysisStats);
  $analysisStats->print_stats;
  
  if($analysisStats->status eq 'BLOCKED') {
    print("Analysis is BLOCKED, can't create workers\n");
    return undef;
  }
  if($analysisStats->status eq 'DONE') {
    print("Analysis is DONE, don't need to create workers\n");
    return undef;
Jessica Severin's avatar
Jessica Severin committed
103
104
105
106
  }

  my $host = hostname;
  my $ppid = getppid;
107
  $beekeeper = '' unless($beekeeper);
Jessica Severin's avatar
Jessica Severin committed
108
109
110
111

  my $sql = "INSERT INTO hive SET born=now(), last_check_in=now()".
            ",process_id='$ppid' ".
            ",analysis_id='$analysis_id' ".
112
            ",beekeeper='$beekeeper' ".
Jessica Severin's avatar
Jessica Severin committed
113
114
115
116
117
118
119
120
121
122
            ",host='$host'";

  my $sth = $self->prepare($sql);
  $sth->execute();
  my $hive_id = $sth->{'mysql_insertid'};
  $sth->finish;

  my $worker = $self->_fetch_by_hive_id($hive_id);
  $worker=undef unless($worker and $worker->analysis);

123
  if($worker and $analysisStats) {
124
    $analStatsDBA->update_status($analysis_id, 'WORKING');
125
    $worker->batch_size($analysisStats->batch_size);
Jessica Severin's avatar
Jessica Severin committed
126
127
128
129
130
131
132
133
134
  }
  return $worker;
}


sub register_worker_death {
  my ($self, $worker) = @_;

  return unless($worker);
135
136
137
138

  # if called without a defined cause_of_death, assume catastrophic failure
  $worker->cause_of_death('FATALITY') unless(defined($worker->cause_of_death));
  
Jessica Severin's avatar
Jessica Severin committed
139
140
141
142
143
144
145
146
  my $sql = "UPDATE hive SET died=now(), last_check_in=now()";
  $sql .= " ,work_done='" . $worker->work_done . "'";
  $sql .= " ,cause_of_death='". $worker->cause_of_death ."'";
  $sql .= " WHERE hive_id='" . $worker->hive_id ."'";

  my $sth = $self->prepare($sql);
  $sth->execute();
  $sth->finish;
147
148

  if($worker->cause_of_death eq "NO_WORK") {
149
    $self->db->get_AnalysisStatsAdaptor->update_status($worker->analysis->dbID, "ALL_CLAIMED");
150
151
152
  }
  if($worker->cause_of_death eq "FATALITY") {
    #print("FATAL DEATH Arrrrgggghhhhhhhh (hive_id=",$worker->hive_id,")\n");
153
    $self->db->get_AnalysisJobAdaptor->reset_dead_jobs_for_worker($worker);
154
  }
Jessica Severin's avatar
Jessica Severin committed
155
156
157
}


158
sub worker_check_in {
Jessica Severin's avatar
Jessica Severin committed
159
160
161
162
163
164
165
166
167
168
169
170
  my ($self, $worker) = @_;

  return unless($worker);
  my $sql = "UPDATE hive SET last_check_in=now()";
  $sql .= " WHERE hive_id='" . $worker->hive_id ."'";

  my $sth = $self->prepare($sql);
  $sth->execute();
  $sth->finish;
}


171
172
173
174
175
176
177
178
179
180
sub fetch_overdue_workers {
  my ($self,$overdue_secs) = @_;

  $overdue_secs = 3600 unless(defined($overdue_secs));

  my $constraint = "h.cause_of_death='' ".
                   "AND (UNIX_TIMESTAMP()-UNIX_TIMESTAMP(h.last_check_in))>$overdue_secs";
  return $self->_generic_fetch($constraint);
}

Jessica Severin's avatar
Jessica Severin committed
181

182
183
184
185
186
187
188
189
sub update_analysis_stats {
  my $self = shift;

  my $sql = "SELECT analysis.analysis_id, status, count(*) ".
            "FROM analysis_job, analysis ".
            "WHERE analysis_job.analysis_id=analysis.analysis_id ".
            "GROUP BY analysis_job.analysis_id, status";

190
  my $statsDBA = $self->db->get_AnalysisStatsAdaptor;
191
192
193
194
195
196
  my $analysisStats = undef;

  my $sth = $self->prepare($sql);
  $sth->execute();
  while (my ($analysis_id, $status, $count)=$sth->fetchrow_array()) {
    unless(defined($analysisStats) and $analysisStats->analysis_id==$analysis_id) {
197
      $analysisStats->determine_status()->update() if($analysisStats);
198
199
200
201
202
203
204
205
206
207
208
209
210
211

      $analysisStats = $statsDBA->fetch_by_analysis_id($analysis_id);
      $analysisStats->total_job_count(0);
      $analysisStats->unclaimed_job_count(0);
      $analysisStats->done_job_count(0);
      $analysisStats->num_required_workers(0);
    }

    my $total = $analysisStats->total_job_count();
    $analysisStats->total_job_count($total + $count);
    
    if($status eq 'READY') {
      $analysisStats->unclaimed_job_count($count);
      my $numWorkers = $count/$analysisStats->batch_size;
212
      $numWorkers=1 if($numWorkers<1);
213
      if($analysisStats->hive_capacity>0 and $numWorkers > $analysisStats->hive_capacity) {
214
215
216
217
218
219
        $numWorkers=$analysisStats->hive_capacity;
      }
      $analysisStats->num_required_workers($numWorkers);
    }
    if($status eq 'DONE') { $analysisStats->done_job_count($count); }
  }
220
221
222
223
224
225
226
227
228
229
  $analysisStats->determine_status()->update() if($analysisStats);
  $sth->finish;

  $self->adjust_stats_for_living_workers();
}


sub adjust_stats_for_living_workers {
  my $self = shift;

230
  my $statsDBA = $self->db->get_AnalysisStatsAdaptor;
231
232
233
234
235
236
  
  my $sql = "SELECT analysis_id, count(*) FROM hive ".
            "WHERE cause_of_death='' GROUP BY analysis_id";
  my $sth = $self->prepare($sql);
  $sth->execute();
  while (my ($analysis_id, $liveCount)=$sth->fetchrow_array()) {
237

238
239
240
241
242
243
244
245
246
247
248
249
250
    my $analysis_stats = $statsDBA->fetch_by_analysis_id($analysis_id);

    if($analysis_stats->hive_capacity > 0) {
      my $numWorkers = $analysis_stats->num_required_workers;

      my $capacityAdjust = ($numWorkers + $liveCount) - $analysis_stats->hive_capacity;
      $numWorkers -= $capacityAdjust if($capacityAdjust > 0);
      $numWorkers=0 if($numWorkers<0);

      $analysis_stats->num_required_workers($numWorkers);
      $analysis_stats->update;
    }
  }
251
252
253
254
  $sth->finish;
}


255
256
257
sub next_clutch {
  my $self = shift;

258
  my $clutches = $self->db->get_AnalysisStatsAdaptor->fetch_by_needed_workers();
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
  return (0, 0) unless($clutches);

  my $smallestClutch = undef;
  
  # keep simple : return the clutch with the smallest number of workers first
  #print("\nCheck for CLUTCH!!\n");
  foreach my $analysis_stats (@{$clutches}) {
    #$analysis_stats->print_stats();

    $smallestClutch = $analysis_stats unless($smallestClutch);
    if($analysis_stats->num_required_workers < $smallestClutch->num_required_workers) {
      $smallestClutch = $analysis_stats;
    }
  }
  
  if($smallestClutch) {
    return ($smallestClutch->analysis_id, $smallestClutch->num_required_workers);
  }
  return (0,0);
}


281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
sub check_blocking_control_rules
{
  my $self = shift;

  my $analysisStatsList = $self->db->get_AnalysisStatsAdaptor->fetch_by_status('BLOCKED');
  foreach my $stats (@{$analysisStatsList}) {
    #print("BLOCKED analysis ");  $stats->print_stats;
    my $ctrlRules = $self->db->get_AnalysisCtrlRuleAdaptor->
                    fetch_by_ctrled_analysis_id($stats->analysis_id);
    my $allRulesDone = 1;                    
    foreach my $ctrlrule (@{$ctrlRules}) {
      #use this method because the condition_analysis objects can be
      #network distributed to a different database so use it's adaptor to get
      #the AnalysisStats object
      my $condStats = $ctrlrule->condition_analysis->stats;
      $allRulesDone = 0 unless($condStats->status eq 'DONE');
      #print("  "); $condStats->print_stats;
    }

    if($allRulesDone and @{$ctrlRules}) {
      #print("  UNBLOCK analysis : all conditions met\n");
      $stats->adaptor->update_status($stats->analysis_id, 'READY');
    }
    
  }
}

308
309
310
311
312
313
314
315
316
317
sub print_hive_status
{
  my $self = shift;

  my $allStats = $self->db->get_AnalysisStatsAdaptor->fetch_all();

  foreach my $analysis_stats (@{$allStats}) {
    $analysis_stats->print_stats;
  }
}
318

319
320


Jessica Severin's avatar
Jessica Severin committed
321
322
323
324
325
326
327
328
329
#
# INTERNAL METHODS
#
###################

=head2 _fetch_by_hive_id

  Arg [1]    : int $id
               the unique database identifier for the feature to be obtained
330
  Example    : $feat = $queen->fetch_by_dbID(1234);
Jessica Severin's avatar
Jessica Severin committed
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
  Description: Returns the feature created from the database defined by the
               the id $id.
  Returntype : Bio::EnsEMBL::Hive::Worker
  Exceptions : thrown if $id is not defined
  Caller     : general

=cut

sub _fetch_by_hive_id{
  my ($self,$id) = @_;

  unless(defined $id) {
    $self->throw("fetch_by_dbID must have an id");
  }

  my @tabs = $self->_tables;

  my $constraint = "h.hive_id = $id";

  #return first element of _generic_fetch list
  my ($obj) = @{$self->_generic_fetch($constraint)};
  return $obj;
}

=head2 _generic_fetch

  Arg [1]    : (optional) string $constraint
               An SQL query constraint (i.e. part of the WHERE clause)
  Arg [2]    : (optional) string $logic_name
               the logic_name of the analysis of the features to obtain
  Example    : $fts = $a->_generic_fetch('contig_id in (1234, 1235)', 'Swall');
  Description: Performs a database fetch and returns feature objects in
               contig coordinates.
  Returntype : listref of Bio::EnsEMBL::SeqFeature in contig coordinates
  Exceptions : none
  Caller     : BaseFeatureAdaptor, ProxyDnaAlignFeatureAdaptor::_generic_fetch

=cut

sub _generic_fetch {
  my ($self, $constraint, $join) = @_;

  my @tables = $self->_tables;
  my $columns = join(', ', $self->_columns());

  if ($join) {
    foreach my $single_join (@{$join}) {
      my ($tablename, $condition, $extra_columns) = @{$single_join};
      if ($tablename && $condition) {
        push @tables, $tablename;

        if($constraint) {
          $constraint .= " AND $condition";
        } else {
          $constraint = " $condition";
        }
      }
      if ($extra_columns) {
        $columns .= ", " . join(', ', @{$extra_columns});
      }
    }
  }

  #construct a nice table string like 'table1 t1, table2 t2'
  my $tablenames = join(', ', map({ join(' ', @$_) } @tables));

  my $sql = "SELECT $columns FROM $tablenames";

  my $default_where = $self->_default_where_clause;
  my $final_clause = $self->_final_clause;

  #append a where clause if it was defined
  if($constraint) {
    $sql .= " WHERE $constraint ";
    if($default_where) {
      $sql .= " AND $default_where ";
    }
  } elsif($default_where) {
    $sql .= " WHERE $default_where ";
  }

  #append additional clauses which may have been defined
  $sql .= " $final_clause";

  my $sth = $self->prepare($sql);
  $sth->execute;

#  print STDERR $sql,"\n";

  return $self->_objs_from_sth($sth);
}

sub _tables {
  my $self = shift;

  return (['hive', 'h']);
}

sub _columns {
  my $self = shift;

  return qw (h.hive_id
             h.analysis_id
434
             h.beekeeper
Jessica Severin's avatar
Jessica Severin committed
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
             h.host
             h.process_id
             h.work_done
             h.born
             h.last_check_in
             h.died
             h.cause_of_death
            );
}

sub _objs_from_sth {
  my ($self, $sth) = @_;

  my %column;
  $sth->bind_columns( \( @column{ @{$sth->{NAME_lc} } } ));

  my @workers = ();

  while ($sth->fetch()) {
    my $worker = new Bio::EnsEMBL::Hive::Worker;
    $worker->init;

    $worker->hive_id($column{'hive_id'});
458
    $worker->beekeeper($column{'beekeeper'});
Jessica Severin's avatar
Jessica Severin committed
459
460
461
462
463
464
465
    $worker->host($column{'host'});
    $worker->process_id($column{'process_id'});
    $worker->work_done($column{'work_done'});
    $worker->born($column{'born'});
    $worker->last_check_in($column{'last_check_in'});
    $worker->died($column{'died'});
    $worker->cause_of_death($column{'cause_of_death'});
466
    $worker->queen($self);
Jessica Severin's avatar
Jessica Severin committed
467
468
    $worker->db($self->db);

469
470
    if($column{'analysis_id'} and $self->db->get_AnalysisAdaptor) {
      $worker->analysis($self->db->get_AnalysisAdaptor->fetch_by_dbID($column{'analysis_id'}));
Jessica Severin's avatar
Jessica Severin committed
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
    }

    push @workers, $worker;
  }
  $sth->finish;

  return \@workers
}

sub _default_where_clause {
  my $self = shift;
  return '';
}

sub _final_clause {
  my $self = shift;
  return '';
}


1;