tailieunhanh - QUERYING AND MANIPULATING TEMPORAL DATABASES

It is increasingly common to find tables with thousands of col- umns – they arise when a particular object has thousands of meas- ured attributes. Not infrequently, many of the values are null. For example, an LDAP object has 7 required and a thousand optional attributes. It is convenient to think of each object as a row of a table, but representing it that way is very inefficient – both in space and. | International Journal of Database Management Systems IJDMS February 2011 Querying and Manipulating Temporal databases Mohamed Mkaouar1 Rafik Bouaziz2 and Mohamed Moalla1 1Université de Tunis El Manar Faculté des Sciences de Tunis Campus Universitaire 2092 - El Manar Tunis Tunisie 2Université de Sfax Faculté des Sciences Economiques et de Gestion de Sfax Route de l Aéroport 3018 Sfax Tunisie Abstract Many works have focused for over twenty five years on the integration of the time dimension in databases DB . However the standard SQL3 does not yet allow easy definition manipulation and querying of temporal DBs. In this paper we study how we can simplify querying and manipulating temporal facts in SQL3 using a model that integrates time in a native manner. To do this we propose new keywords and syntax to define different temporal versions for many relational operators and functions used in SQL. It then becomes possible to perform various queries and updates appropriate to temporal facts. We illustrate the use of these proposals on many examples from a real application. Keywords Temporal Databases Querying and manipulating temporal facts SQL3 extension. 1. Introduction Works on temporal databases temporal DB 1 2 3 aim the modelling and the manipulation of different types of temporal facts based on two kinds of time valid-time and transaction-time 4 . By Timestamping facts with one or both kinds of these times we distinguish three types of temporal facts valid-time facts transaction-time facts and bitemporal facts. Valid-time facts may relate to the past to the present or to the future. Each such fact is represented by a timestamp with their valid-times in reality. These times may be instants time intervals or temporal elements defined in accordance with a given calendar and a given granularity. Thus we can maintain the history of all valid facts which can be updated in real time .