Gaffar, Ashraf (2001) Design of a framework for database indexes. Masters thesis, Concordia University.
Database system performance depends greatly on the performance of the indexes used to lookup and update the database, therefore it is important to have efficient indexes to the database. Specialized application indexes developed by experts have "specialized source code" for each kind of database application. The time and cost to develop an index specific to the kind of application could be very high, making it unaffordable or even unavailable in many cases. Object-oriented framework technology has been used to produce index frameworks that can be applied to develop indexes, reducing the development cost. An index framework allows one to adapt to different key/data types, different queries and different access methods. In this thesis, we focus on balanced tree indexes, and develop a framework in the style of the STL. We focus on the early stages of analysis, architecture, and design in an object-oriented methodology in order to design the framework. We achieve a modular framework with decoupled modules for data, data references, containers, indexes, iterators, and algorithms. This allows for developing new applications by replacing some of these modules and reflecting the changes from one model to the next, without affecting the other modules. This would result in easier developing process with less steep learning curve, and produces applications that have their own "specialized" architecture, design and source code.
|Divisions:||Concordia University > Faculty of Engineering and Computer Science > Computer Science and Software Engineering|
|Item Type:||Thesis (Masters)|
|Pagination:||xi, 130 leaves : ill. ; 29 cm.|
|Degree Name:||Theses (M.Comp.Sc.)|
|Program:||Computer Science and Software Engineering|
|Thesis Supervisor(s):||Butler, Gregory|
|Deposited By:||Concordia University Libraries|
|Deposited On:||27 Aug 2009 17:19|
|Last Modified:||08 Dec 2010 15:20|
Repository Staff Only: item control page
Downloads per month over past year