# TDDD07 Real Time Systems

### Bonus Assignments

The written examination can be complemented by preparing answers to bonus assignments in advance. The purpose of these assignments is to ensure that students that aim for deeper understanding of the material in the course have additional material to work with. They are intended for students who are not finding the rest of the material as challenging enough.
The assignments will be introduced during the course, and a student (optionally) hands in the individually prepared answers to the examiner. Please send your analysis reports in pdf (max 2 pages) to the examiner (simin.nadjm-tehrani@liu.se), and in the document provide your LiU ID and bonus assignment #.
Please note that you are allowed to consult books and other Internet resources, but the answers to the assignments #1 and #2 are to be formulated entirely by yourself and can be subject to a verbal discussion. The deadline for the submission is 13 December so that the examiner can give you feedback before the end of the year.
For assignment #3 which is related to the Rescue lab application, you include your analysis text in an individual report (see also the lab compendium). Note that assessment for this assignment is on an individual basis. You need to demonstrate your code to the lab assistant (in a slot booked on demand) and hand in your report before 16 December.

Bonus assignment #1: How are the proofs for optimality of RMS in the Buttazzo book (chapter 4) affected by the assumptions on the task set (that are listed in that chapter)?
I expect a description that includes a couple of sentences in your own words for each of the assumptions, that shows me you have understood where in the proof there is a need for each assumption. Thus, you can formulate a clear reason for why those assumptions are needed.
This assignement can give you 2 \$B points.

Bonus assignment #2: To get a deeper understanding of the differences between RMS and EDF scheduling you are asked to read the article by Buttazzo (2005) and use your own words to answer the following questions:
• Analysing the merits of the two algorithms, is there a conclusive outcome? How would you choose among them? Why?
• How does the paper justify the propostion that "EDF introduces less run-time overhead compared to RMS"? Do you have any doubts about validity of this claim?
This assignement can give you 2 \$B points.

Bonus assignment #3: Fault mode classifications as described in the Avizienis et al. article in the course literature is a tool for concretising how a system developer tackles the foreseen faults, errors, failures in a system at design stage. Some fault classes (modes) can/should only be treated through prevention, some through removal and some through tolerance. The purpose of this assignment is to test your deeper understanding of how these classifications can be a tool in a real scenario. You are expected to consider the rescue robot scenario that you have become familiar with by the end of Lab 2 and decide if there are any fault modes that now, after a partial realisation of the system, are plausible, and whether additional code can be used to treat the fault as this (operational) stage. The addition of this analysis, if supported by valid reasoning, can give you 2 \$B points for providing new pseudocode and, an additional 2\$B points for implementing and demonstrating it.