GitLab

schema_change: removed analysis_stats[_monitor].num_required_workers, substituted with AS::estimate_num_required_workers()

simplified the apparent control flow of synchronization subroutines, removed the returned value that is no longer used

Moved the caching from Hive.pm into the specific storable classes themselves (via Cacheable.pm) to better conform to Active Record pattern

computational part of sync() moved into AnalysisStats::recalculate_from_job_counts and can be run separately

no need to update only the status here, as we are going to update the whole AnalysisStats object very soon.

check_blocking_control_rules() no longer talks to the adaptor, it gets its rules from analysis->control_rules_collection

Switched to our internal implementation of throw() via Carp::confess. It is also available to Runnables as $self->throw( $msg );

Storable->new() now performs setting of all parameters without the use of rearrange (NB: change of call format!)

removed dependency on Core's DBSQL::BaseAdaptor and Storable by introducing own lightweight Hive::Storable

Back to num_required_workers' meaning "how many extra workers we need to add to this analysis"; fixing a scheduling bug/oversensitivity to manual change of batch_size

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 )

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

clearer display of job_counters on the graph; removed misleading and unused remaining_job_count() and cpu_minutes_remaining()