ADVANCED DATA MANAGEMENT

The aim of the course is to present the internals of database management systems and to describe the principal DBMS families. The goal is to enhance the comprehension and the advanced use of these systems by database administrators and database application's programmers.

Ability to choose the most suitable type of DBMS to meet the application requirements and to discuss the effects of architectural, modeling, and configuration choices. Ability to read and understand scientific literature regarding algorithms used in DBMS.

Databases: conceptual and relational data modelling. The query language SQL and the relational algebra

0. Persistent memory and buffer management; File structures in secondary memory; Access methods management in DBMSs; Transaction management; Concurrency in DBMSs; DBMS implementations based on Physical operators.
1. Relational DBMS architectures: Physical operators; Query optimization techniques; Physical design, tuning, and maintenance.
2. Not Only SQL: Graph Databases; XML Databases; Key-Value Stores; Document Databases; Column Stores; Extensible Record Stores; Data Stream Management.
3. Key-value store architectures and support for graphs and documents.
4. Native graph DBMS architectures.
5. Distributed Database Systems: Data Fragmentation, Replication and Synchronization, Consistency
6. Other technologies: Linked Data and SPARQL, Data Stream Management, Geographic Information Systems, In-Memory Databases, NewSQL.

A. Albano, D. Colazzo, G. Ghelli, R. Orsini. DBMS Internals, pag. 1-229, forthcoming book.
L. Wiese. Advanced Data Management for SQL, NOSQL, Cloud and Distributed Databases. 2015. De Gruyter Oldenbourg. ISBN: 978-3-11-044140-6.
Additional material in Moodle.

The examination consists of an oral test that assesses the knowledge of the topics presented during the course and the ability to apply them to real problems in a project or to explore them in depth by analyzing scientific articles.

IMPORTANT: It is mandatory to contact the teacher before the oral exam to agree on the scientific article to study or the requirement of the scenario to be modeled and the software to be implemented.

Regarding grade grading (how grades will be assigned):

A. scores in the 18-22 range will be awarded for:
A1. Basic knowledge of main topics (DBMS families and their characteristics, file structures for data persistence in secondary memory, physical operators, access cost estimation).
A2. One of the following:
A2.1. the ability to use a NoSQL DBMS to model a simple use case, partially implement a working software for some of its functionalities, and appropriately discuss design and implementation choices (project prepared before the exam, ability to repeat individual steps).
A2.1. the ability to read, understand, and present a scientific publication and to identify its main contributions (presentation prepared before the exam, ability to relate part of the presentation to the corresponding part of the paper).

B. scores in the 23-27 range will be awarded for:
B1. As for A1, with a thorough knowledge of the main topics and good knowledge of the other course topics.
B2. One of the following:
B2.1. As for A2.1, for a non-trivial use case plus the ability to discuss part of the functionalities of the chosen DBMS in detail and their use in other hypothetical use cases.
B2.2. As for A2.2, plus good presentation and ability to relate the contributions of the scientific paper to the course topics and discuss its main limitations.

C. scores in the 28-30L range will be awarded in the presence of:
C1. As for B1, with a thorough knowledge of all the course topics
C2. One of the following:
C2.1. As for B2.1, plus an excellent presentation of the activities and in-depth knowledge of the chosen DBMS.
C2.2. As for B2.2, plus excellent presentation and in-depth understanding of the scientific article's proposal, limits, and contribution.

Lectures and lab sessions.

English

oral