News

  Tweets by @pdrtschalmers

Staff (contact information)

• Jan Jonsson, Lecturer and Examiner
• Risat Pathan, Lecturer
• Behrooz Sangchoolie, Course assistant

Course elements

• Sixteen lectures.
• Two homework assignments, done in groups of at most two people.
• A voluntary written exam.

Aim and context

Real-time systems play a vital role in many application domains including drive-by-wire automotive systems, intelligent traffic control systems, banking systems, and autonomous robots. For many real-time applications, the system must be designed with multiple processors in order to meet imposed application constraints. Such situations can occur when tasks need to be physically distributed (as in a fault-tolerant vehicle braking system) or when the inherent parallelism of the application need to be fully exploited (as in a high-performance multimedia application).

After the course the students shall be able to:

• Formulate requirements for computer systems used in time- and safety critical applications.
• Master the terminology of scheduling and complexity theory.
• Describe the principles and mechanisms used for scheduling of task execution and data communication in real-time systems.
• Derive performance for, and be familiar with the theoretical performance limitations of, a given real-time system.

This course is intended to give a deeper understanding of the problems involved in designing real-time systems based on multiprocessor architectures, and covers the following topics:

• Background: motivation for, and definition of, real-time computing systems.
• Characteristics of real-time systems: application constraints, design methods, task models, run-time mechanisms, architectures.
• Evaluation of real-time systems: performance measures, evaluation methodologies.
• Single and multiprocessor scheduling: problem definition, terminology, and algorithms.
• Complexity theory and NP-completeness in the context of real-time scheduling.
• Real-time communications: protocols and end-to-end delay guarantees.
• Fault-tolerance techniques for real-time systems: models, algorithms and architectures.
• Distributed clock synchronization.
• Estimation of task execution times.

Context

The course requires a BSc in Computer Science and Engineering or equivalent. Preparatory courses in real-time systems, algorithms, operating systems and computer architecture are desired but not required.

Important Dates

	Mon	Mar 23	:	First lecture, 13.15 - 15.00 in ES52
	Mon	Apr 20	:	Homework assignment #1 handed out
	Fri	May 8	:	Deadline, Homework assignment #1
	Fri	May 8	:	Homework assignment #2 handed out
	Tue	May 26	:	Deadline, Homework assignment #2
	Thu	June 4	:	Voluntary written exam, afternoon, Johanneberg

Lectures

The course is organized as a series of lectures where fundamental theories and concepts are presented. Lectures are are offered on the following time and study weeks:

	Mon		13.15 - 15.00		ES52		(week 1, 2, 3, 4, 5 and 6)
	Thu		08.00 - 09.45		ES52		(week 2 and 7)
	Thu		10.00 - 11.45		ES52		(week 1, 2, 3, 4, 7 and 8)
	Fri		15.15 - 17.00		ES52		(week 1 and 8)

As a complement to the lectures, there will be a homework assignment on the specific topics covered during the scheduled lectures. As an aid for the students for their homework, consultation sessions will be offered. The consultation sessions offer assistance regarding questions and problems related to the homework assignment (or other related issues). The intention is for the student to attempt to solve the problems and then go to the consultation session in case help is needed. Consultation sessions are offered on the following time and study weeks:

	Mon		13.15 - 15.00		ES52		(week 7 and 8)
	Thu		08.00 - 09.45		ES52		(week 4 and 5)
	Thu		10.00 - 11.45		ES52		(week 5)
	Fri		15.15 - 17.00		ES52		(week 3 and 7)

Preliminary schedule: TimeEdit

Detailed information on the lectures will appear here.

Course literature

Selected research publications from journals and conference proceedings (downloadable here.)

Course Evaluation

For the purpose of course evaluation, we ask volunteers to act as student representatives. Their role includes giving the teachers some feedback on the course. Please send them an email if you have comments or suggestions for improvements regarding the course.

The student representatives are:

	Johan Gustafsson (MPCSN)
	Fredrik Hidstrand (MPCSN)
	Henrik Hugo (MPCSN)
	Emil Lindqvist (MPEES)
	Eyþór Sigmundsson (MPCSN)

Minutes from mid-quarter course evaluation meeting can be found here.

Chalmers central instructions on course evaluation