Dear developing team, 




We are working on an improvement of the
job scheduler currently implemented in Bacula and we are wondering if
you could be interested in our work..





The new features of the scheduler
are the following:





        
        The FCFS algorithm used in Bacula
        is replaced by a dynamic strategy.
        
                At
                 run
                time,
                a dynamic priority value is associated to each job.
                An
                aging
                process is in charge of periodically updating the dynamic 
priority
                of each job, increasing the importance of the job as time goes 
by. 
                
                
                The job with the highest dynamic
                priority is the first to run.
        






        
        New job behaviors can be defined
        
                A job can run at regular
                intervals with respect to last instance’s 
                
                
                        planned start time,
                        actual start time,
                        
                        completion  time.
                
                
                Time windows can be defined for
                each job in which the job must not be started by the scheduler.
                
                Preferred time windows can be
                set for a job’s start time.
                A
                job can be configured to run
                before its planned start time if
                the system is under specific conditions
        






        A
        recovery
        strategy can
        be enabled:
        
                
                In case of a system failure
                (e.g. after a shutdown of the backup server for maintenance)
                
                        
                        The scheduler is able to
                        recalculate the expected schedule as if the system did 
not fail.
                        
                        This means that jobs that have
                        not run for a long period, are likely to be run first. 
                        
                
                
                In case of a job failure (e.g.
                the node to be backed up is temporarily unavailable)
                
                        The job can
                        be rescheduled again after a configurable delay.
                
        





We want to point out that our work
can be related to item 14 and item 15 of the Bacula Project Roadmap.



Who should we
contact for a deeper discussion on the subject?
Kind Regards,




Paolo Di Francesco,
        paolo_p...@hotmail.com  

Stefano Stalio, 
        stefano.sta...@lngs.infn.it                                       
------------------------------------------------------------------------------
Cloud Services Checklist: Pricing and Packaging Optimization
This white paper is intended to serve as a reference, checklist and point of 
discussion for anyone considering optimizing the pricing and packaging model 
of a cloud services business. Read Now!
http://www.accelacomm.com/jaw/sfnl/114/51491232/
_______________________________________________
Bacula-devel mailing list
Bacula-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-devel

Reply via email to