Commit c22c76e5 authored by Leo Gordon's avatar Leo Gordon
Browse files

cosmetic: added new commits to README and renamed it Changelog; split out the...

cosmetic:  added new commits to README and renamed it Changelog; split out the old README.txt (non-Changelog part)
parent b57d7d04
# Bio::EnsEMBL::Hive project
#
# Copyright Team Ensembl
# You may distribute this package under the same terms as perl itself
Contact:
Please contact ehive-users@ebi.ac.uk mailing list with questions/suggestions.
Summary:
This is a distributed processing system based on 'autonomous agents' and
Hive behavioural structure of Honey Bees. It implements all functionality of both
data-flow graphs and block-branch diagrams which should allow it to codify
any program, algorithm, or parallel processing job control system. It is
not bound to any processing 'farm' system and can be adapted to any GRID.
Changelog:
---------------------[before and during EnsEMBL rel.71]----------------------------------------------------------------
* Thu Feb 28 10:12:41 2013 +0000 | Leo Gordon | avoid having beekeeper run in submitted-to-the-farm state - detect it, report and quit
* Thu Feb 28 09:47:40 2013 +0000 | Leo Gordon | param_substitution is now default everywhere, no need to call it explicitly
* Thu Feb 28 09:42:33 2013 +0000 | Leo Gordon | added param_required() and param_is_defined() interfaces to Process
* Wed Feb 27 21:34:47 2013 +0000 | Leo Gordon | bugfix: updated examples of how to use JobFactory without and with input_id_template
* Wed Feb 27 19:08:40 2013 +0000 | Leo Gordon | bugfix: changed implementation of data_dbc() to correctly compare things before caching
* Wed Feb 27 14:00:42 2013 +0000 | Leo Gordon | Clone::clone is no longer used, so dependency has been removed
* Fri Feb 22 16:55:12 2013 +0000 | Matthieu Muffato | It is more efficient to give MySQL a LIMIT clause
* Sat Feb 23 00:52:57 2013 +0000 | Leo Gordon | JobFactory uses $overriding_hash to create jobs/rows from input_id_template; 'input_id' parameter deprecated; standaloneJob supports templates.
* Sat Feb 23 00:49:15 2013 +0000 | Leo Gordon | Substitution machinery now supports an extra $overriding_hash that contains parameters with higher precedence than the whole of param() structure
* Fri Feb 22 16:36:19 2013 +0000 | Leo Gordon | fixed several problems with parameter substitution and detection of undefs; added param_required() and param_is_defined()
* Fri Feb 22 10:42:51 2013 +0000 | Leo Gordon | reload the cached data_dbc() value on change of param('db_conn')
* Thu Feb 21 16:14:35 2013 +0000 | emepyc | The modules of the analyses must be accessible
* Fri Feb 15 17:05:20 2013 +0000 | Matthieu Muffato | New runnable to check the size of the resultset of any SQL query
* Tue Feb 19 17:18:06 2013 +0000 | Leo Gordon | removed param_substitute() call from Runnables -- no longer needed, as substitution is automatic
* Tue Feb 19 16:46:05 2013 +0000 | Leo Gordon | a "total" (anything-to-anything) substitution mechanism has been implemented in Hive::Params
* Fri Feb 15 17:04:36 2013 +0000 | Matthieu Muffato | The preferred meadow type must be registered
* Fri Feb 15 17:03:24 2013 +0000 | Matthieu Muffato | The modules of the analysis must be loadable
* Fri Feb 15 17:01:27 2013 +0000 | Matthieu Muffato | In dataflow rules within the same database, the destination analysis must exist
* Fri Feb 15 17:00:38 2013 +0000 | Matthieu Muffato | In control rules within the same database, the condition analysis must exist
* Fri Feb 15 22:25:39 2013 +0000 | Leo Gordon | make sure all LSF pids are quoted, to protect them from tcsh interpretation of square brackets
* Thu Feb 14 16:41:49 2013 +0000 | Leo Gordon | a new script to remove old "DONE" jobs and associated job_file and log_message entries
* Thu Feb 14 10:45:26 2013 +0000 | Leo Gordon | seed_pipeline.pl now shows examples of input_ids of seedable analyses
* Thu Feb 14 09:54:00 2013 +0000 | Leo Gordon | Made $final_clause a parameter of _generic_fetch() & removed default ORDER-BY; hopefully faster
* Wed Feb 13 17:20:14 2013 +0000 | Leo Gordon | In case no -logic_name/-analysis_id was supplied, show the list of analyses that have no incoming dataflow (and so are candidates for seeding)
* Wed Feb 13 13:52:56 2013 +0000 | Leo Gordon | Added perldoc to seed_pipeline.pl script
* Wed Feb 13 13:35:55 2013 +0000 | Leo Gordon | A new script to quickly seed any analysis of any pipeline.
* Wed Feb 13 10:19:08 2013 +0000 | Leo Gordon | hide the calls to URLFactory into the DBAdaptor's constructor
* Tue Feb 12 10:22:02 2013 +0000 | Leo Gordon | hash of resources no longer depends on default_meadow (bugfix)
* Thu Feb 7 11:42:11 2013 +0000 | Kathryn Beal | Updated to release 71
* Wed Feb 6 17:43:21 2013 +0000 | Matthieu Muffato | Tables must be in the right order. Otherwise, the foreign key checks complain
* Fri Jan 25 19:42:28 2013 +0000 | Leo Gordon | resolving conflict: using mine
|\
| * Tue Jan 15 11:03:26 2013 +0000 | Matthieu Muffato | Table dataflows are now included into semaphore boxes (bugfix: wrong internal name)
* | Fri Jan 25 19:26:36 2013 +0000 | Leo Gordon | diagram improvement: (1) no more "empty boxes" and (2) tables dataflown from a box are shown in their boxes
|/
* Mon Jan 14 13:23:52 2013 +0000 | Leo Gordon | Added a new presentation, moved presentations into a separate folder.
* Fri Jan 11 11:19:11 2013 +0000 | Leo Gordon | cosmetic fix: commented back the debug output that was left uncommented by mistake
* Fri Jan 11 11:07:47 2013 +0000 | Leo Gordon | Added coloured barchart display option and jobs/data display option (no big data checks, use with care on small examples). 'Pad' is now configurable from JSON. Beware: JSON config options have moved around!
* Thu Jan 10 16:14:06 2013 +0000 | Leo Gordon | injected a padding around the pipeline diagram
* Fri Jan 4 17:03:14 2013 +0000 | Leo Gordon | send the fatal "COULDNT CREATE WORKER" message to stderr instead of stdout
* Fri Jan 4 15:10:47 2013 +0000 | Leo Gordon | added command line options -submit_stdout_file and -submit_stderr_file to peek into submission output/error streams
* Fri Jan 4 14:51:35 2013 +0000 | Leo Gordon | using PERLs File::Path::make_path instead of mkdir-p to create hive_log_dir
* Fri Jan 4 11:03:31 2013 +0000 | Leo Gordon | added a LongMult pipeline diagram in completed state (for easier reference)
* Mon Dec 17 12:13:43 2012 +0000 | Leo Gordon | fixed a bug in computing num_required_workers according to the new rules (thanks Matthieu for reporting)
* Wed Dec 12 14:41:16 2012 +0000 | Leo Gordon | bugfix: correctly checking analysis_capacity (thanks Andy for reporting)
* Wed Dec 12 10:44:01 2012 +0000 | Leo Gordon | bugfix: do not proceed with negative numbers of workers for submission (thanks to Matthieu for reporting)
* Thu Dec 6 11:18:59 2012 +0000 | Leo Gordon | bugfix: -job_limit now works correctly also with respecializing workers
* Thu Dec 6 10:47:07 2012 +0000 | Leo Gordon | fix: 'msg' view now displays the analysis of the job (which is fixed), not that of worker (which may change with time)
* Wed Dec 5 22:25:35 2012 +0000 | Leo Gordon | experimental feature: re-specialization of workers instead of dying from NO_WORK
* Sat Dec 1 19:11:56 2012 +0000 | Leo Gordon | switched to using Limiter class for job_limit and made some related structural changes
* Fri Nov 30 13:47:42 2012 +0000 | Leo Gordon | changed the meaning of 'num_required_workers' to "total estimated number of workers needed for this analysis" ( 'num_running_workers' is now included in it )
* Thu Nov 29 12:21:22 2012 +0000 | Leo Gordon | fresh schema diagram
* Thu Nov 29 11:46:47 2012 +0000 | Leo Gordon | renamed 'job_message' table to 'log_message' and JobMessageAdaptor to LogMessageAdaptor everywhere
* Wed Nov 28 21:40:45 2012 +0000 | Leo Gordon | swapped hive_capacity for analysis_capacity in example PipeConfig files
* Wed Nov 28 21:30:44 2012 +0000 | Leo Gordon | change of default behaviour: hive_capacity is now off by default (=NULL); setting hive_capacity=0 or analysis_capacity=0 stops scheduling AND specialization to a particular analysis
* Wed Nov 28 13:23:48 2012 +0000 | Leo Gordon | cleanup: removed runnable(), output() and parameters() subroutines from Process as no longer used by Compara
* Wed Nov 28 12:21:37 2012 +0000 | Leo Gordon | removed the "compile_module_once" option as the only way to compile modules now is once after specialization
* Tue Nov 27 11:31:00 2012 +0000 | Leo Gordon | secutiry: make sure stringify() always produces perl-parsable structures, so that global settings of Data::Dumper do not affect its results (thanks to Uma and Matthieu for reporting)
---------------------[during EnsEMBL rel.70]----------------------------------------------------------------
......
# Bio::EnsEMBL::Hive project
#
# Copyright Team Ensembl
# You may distribute this package under the same terms as Perl itself
Summary:
Ensembl Hive is a distributed processing system based on 'autonomous agents' and
Hive behavioural structure of Honey Bees. It implements all functionality of both
data-flow graphs and block-branch diagrams which should allow it to codify
any program, algorithm, or parallel processing job control system.
It is not bound to any processing 'farm' system and can be adapted to any GRID.
Contact:
If you have any questions/suggestions, please use our dedicated mailing list.
To subscribe please visit:
http://listserver.ebi.ac.uk/mailman/listinfo/ehive-users
After subscribing you will be able to post to:
ehive-users@ebi.ac.uk
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment