Persistent applications using generalized redo recovery

摘要: We describe how to recover applications after system crashes using database recovery. Earlier efforts, based on frequent application checkpoints and/or logging values read, are very expensive. treat state as a cached object and log execution operations in the recovery framework of D. Lomet M. Tuttle (1995). Logging does not require state. Further logged reads mostly logical which only data source identity is logged. cache manager that handles flush order dependencies introduced by these process restores replaying application.