IT-Universitetet i København
 
  Tilbage Kursusoversigt
Kursusbeskrivelse
Kursusnavn (dansk):Database Tuning 
Kursusnavn (engelsk):Database Tuning 
Semester:Forår 2016 
Udbydes under:cand.it., softwareudvikling og -teknologi (sdt) 
Omfang i ECTS:7,50 
Kursussprog:Engelsk 
Kursushjemmeside:https://learnit.itu.dk 
Min. antal deltagere:12 
Forventet antal deltagere:40 
Maks. antal deltagere:40 
Formelle forudsætninger:In order to participate at this course you must have completed the course Database Systems (DBS). or have achieved equivalent competencies elsewhere.
-----
Information about the course of study
This course is part of a specialization in Databases on the Master of Science in IT, study programme Software Development and Technology - AC track 
Læringsmål:After the course the student should be able to:

  • Analyze the performance impact of design choices within a database system (indexing, query optimization, concurrency control and recovery)

  • Analyze and Compare the characteristics of different data models (relational, sequence, array)

  • Analyze and Compare the characteristics of the storage architecture (e.g., shared disk, shared nothing)

  • Design and execute experiments to evaluate a data management system performance

 
Fagligt indhold:After decades of development, the software behind relational databases (RDBMSs) makes it possible to design, in a technically relatively simple way, robust software systems that work correctly even in demanding environments with many users, system breakdowns, etc. Often, uses of transactional database systems put high demands on the speed and throughput of the system: Long waiting times are not acceptable, certain types of maintenance and data analysis must run overnight, etc. The goal of this course is to provide students with conceptual and practical tools for improving the performance of relational databases (perform "tuning").

Successful candidates will acquire a thorough understanding of the internals of data management systems; they will be able to choose a data management system that matches given needs and they will be able to tune a data management system for performance.

Reasoning about the performance of a database system requires (a) understanding the internal workings of the database system, and (b) the ability to experiment to either troubleshoot a performance problem or expose the performance characteristics of a given system.

The course deals with the following major topics:
  • Architectures: parallel, main-memory relational database systems
  • Data models: relational, array-based, sequence
  • Physical Data Organization and Indexing
  • Transaction management: Concurrency Control and recovery
  • Query optimization


This course is a requirement for the Database Tuning Project (SDT2). 
Læringsaktiviteter:14 ugers undervisning bestående af forelæsninger og øvelser

Teaching consists of a mix of lectures and exercises. The exercises are in the form of tuning experiments. The written exam will have the same form as the exercises .
----
The schedule will be available shortly before the beginning of the term.
See the schedule here:
link to the time table
-----
NB!! Course restriction!!
Please note that there is a course restriction between this course and the SDT courses Database Tuning, Scientific Data Management and Advanced Data Management.
That means that you cannot take this course, if you have already taken Database Tuning, Advanced Data Management and Scientific Data Management, and that you cannot take Scientific Data Management if you take this course.

In order to complete the Database specialization on SDT, you need to take the Database Tuning Project (SDT2) and Big Data Management (SBDM) 

Obligatoriske aktivititer:Three assignments are given in lectures 4, 6 and 8.
Two of them are mandatory. They are due on May 2nd.
Mandatory assignments can be resubmitted 17th May before the ordinary exam. 
Eksamensform og -beskrivelse:A11: Skriftlig eksamen (stedprøve) med adgang til internet, skriftlige og trykte hjælpemidler., (7-scale, external exam)

Duration of exam: 4 hours  

Litteratur udover forskningsartikler:The course relies on notes and research papers that will be distributed in class.