Software Systems
Spring 2005

For today, you should have:

1) Prepared for a quiz.

2) Continued wok on Homework 4

3) Read Tanenbaum pages 234-242 and answered the reading questions.


Outline:

1) quiz

2) Proposal feedback

3) Homework 4

4) reading questions


For next time you should:

1) Finish Homework 4

2) Work on your project.

3) Read about memory management.


Homework 4
----------

Code review...

Why are the actual values different from the predicted values?

1) warm up and cool off

2) actual workload vs. nominal workload

3) randomness


How to compute time averages.

1) instrument handle_event

2) use PASTA and add observation events


Reading questions
-----------------

Standish pages 224 -- 237.

1) In general (not just in Java) what kind of variables get
   allocated in the static segment, the stack, and the heap?






The mark and gather algorithm on page 227 is one of the
two basic garbage collection algorithms.  The other, reference
counting, is coming up on page 235.

The example on page 228 is more specific than we care about.
Don't spend too much time on it.

2) On page 230, you might wonder why the available blocks are
   kept in a two-way linked list, but at this point you probably
   don't know.  At the end of the chapter, come back and write
   the reason below...







3) What is the primary benefit of the First-Fit strategy?






4) What is the primary benefit of the Best-Fit strategy?





5) What is the difference between coalescing and compacting?





6) Why is compacting hard to implement?





7) What is double-dereferencing?





8) What is the primary advantage of reference counting over
   mark and collect?





9) What is the primary disadvantage?






10) Please answer question 2 in section 7.6 on page 237.