Commit 81e69e1c authored by Leo Gordon's avatar Leo Gordon
Browse files

listed the major features/changes that went into version/2.1

parent 35e5251f
Version 2.1 - 'multirole'
* Improved internal API that allows implicit lazy-loading of objects associated with other objects via their dbIDs
* Objects that make up pipeline's graph can be loaded into cache, which simplifies structural topup of existing pipeline databases
* Diagram-drawing engine was stripped of its' dependence on dbIDs, so diagrams can now be built directly from PipeConfig file(s) using -pipeconfig option(s)
* -analysis_topup removed (became the default mode of operation), -job_topup removed in favour of providing same functionality
* pipeline_wide_parameters moved into a separate table, so hive-specific 'meta' table is no longer needed, and Ensembl's version can happily coexist
* 'monitor' table removed in favour of offline script (that does not require a constantly running for data generation)
* pipeline_create_commands() is executed even on topup; redefine to return an empty list or use -hive_no_init if you don't need commands to be executed
* Switched to 'worker_resource_usage' table, unified resource collection calls for other Meadows, so SGE/CONDOR/etc resources can be shown in guiHive & timeline.
* Introduced 'role' table and Role objects to better track role-switching of multirole Workers
* Added Process::complete_early() as the blessed way to exit the code early successfully and store a log_message
* More careful semaphore rebalancing strategy that can also be switched on or off during pipeline database generation
* Logging and error reporting has been improved and simplified
* Multiple bugs have been fixed
Version 2.0 - a major 'coreless' release of Hive code
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