Computer Architecture
Datorarkitektur

EITF20F, 7.5 credits

Valid from: Autumn 2017
Decided by: Professor Thomas Johansson
Date of establishment: 2018-03-02

General Information

Division: Electrical and Information Technology
Course type: Course given jointly for second and third cycle
The course is also given at second-cycle level with course code: EITF20
Teaching language: English

Aim

The goal of this course is to introduce the operation of computer systems at the level of Instruction Set Architectures (ISA). It aims to provide a basic understanding of the design principles that govern modern computer architectures and their components. Especially it pays attention to (super scalar) pipelining and memory hierarchy techniques including caches. Implementation and efficiency issues are exemplified. Metrical performance analysis methods are discussed to evaluate architectural alternatives.

Goals

Knowledge and Understanding

For a passing grade the doctoral student must

• - understand a computer architecture
• - know how to quantify the potential performance of a computer
• - have an overview of the architectural alternatives and technological restrictions to solve a computing problem.

Competences and Skills

For a passing grade the doctoral student must

• · Be skilled in establishing computer functionality
• · Be well-versed in system simulation
• · Be experienced in architecture quantification
• · Have an operational knowledge of architectures

Judgement and Approach

For a passing grade the doctoral student must have learnt trade-offs in creating computer support for solving problems

Course Contents

The course introduces the foundations of low-level computer functioning. The course syllabus includes, among other things, the main parts of computer system and low-level programming techniques and the techniques of RISC-processors and pipelining, cache memory and virtual memory. The course also illuminates the alternative design principles of modern computer architectures in order to provide an understanding of their impact on performance. Quantitative methods to evaluate design principles for performance constitutes an important subject of the course. In coarse overview the lectures treat: · Performance · ISA principles · Pipelining · Memory Systems - I/O, Multiprocessor

Course Literature

Hennesy, J. & Patterson, D.: Computer Architecture – A Quantitative Approach. ISBN 9780123704900.

Instruction Details

Types of instruction: Lectures, laboratory exercises

Examination Details

Examination formats: Written exam, oral exam
Grading scale: Failed, pass
Examiner:

Admission Details

Further Information

The course is based on EITF20 Course Coordinator: Liang Liu, liang.liu@eit.lth.se

Course Occasion Information

Contact and Other Information

Course coordinators:
Web page: http://www.eit.lth.se/kurs/eitf20