Splitting up HHH-3608 - DB sequence numbers are not unique when using the pooled SequenceStyleGenerator in multiple JVMs with the same DB Complete into its individual tasks. This part deals with the assumption made in PooledOptimizer that the sequence value is the upper boundary of the current range of values. This is an assumption and the other assumption that the value is the lower boundary is just as valid, so provide for both by adding a new optimizer. The new optimizer class will be org.hibernate.id.enhanced.OptimizerFactory.PooledLoOptimizer and its short name will be "pooled-lo". The already existing one will remain the default, but I will add a config option to allow users to specify which to use per setup ('hibernate.id.prefer_pooling_lo_values').