Temporal Flower Index Eliminating Impact of High Water Mark

Abstract

A huge amount of data is produced daily, which should be managed, handled and stored in the database. Intelligent system characteristics are mostly based on storing data over the time with regards to validity. The core of the system is therefore framed by the temporal database, which offers the possibility for data analysis, decision making or creating future prognoses. None of the data should be stored indefinitely, effective data management is an inevitable part. This paper references historical background and temporal evolution with emphasis on various granularity modeling and managing describes index structure types used in database approaches covering access paths taken out by optimizer techniques. The main contribution of this paper is Flower Index Approach, which aim is to eliminate the impact of database High Water Mark if Full Table Scan access method is used. Thanks to that, we can optimize costs, reduce processing time and increase performance, which is also supported by experiment comparisons.