Bell Labs: Difference between revisions
No edit summary |
|||
(2 intermediate revisions by one other user not shown) | |||
Line 35: | Line 35: | ||
Everyone else will be graded on the quality of the feedback given to the presenters. I expect everyone not presenting to give at least one constructive comment during class - that will get you a B. More participation, the better the grade - unless you start going overboard. Remember, time will be limited! | Everyone else will be graded on the quality of the feedback given to the presenters. I expect everyone not presenting to give at least one constructive comment during class - that will get you a B. More participation, the better the grade - unless you start going overboard. Remember, time will be limited! | ||
==Presentations== | |||
Presentation slides are [http://homeostasis.scs.carleton.ca/~soma/distos/2008-03-03/presentations/ here]. |
Latest revision as of 18:46, 5 March 2008
Readings
Dennis M. Ritchie and Ken Thompson, "The UNIX Time-Sharing System"
Rob Pike et al., "Plan 9 from Bell Labs"
Sean Dorward et al., "The Inferno Operating System"
Rob Pike and Dennis M. Ritchie, "The Styx Architecture for Distributed Systems"
Questions on the Readings
Note that on Monday there will be a general group discussion on the readings - we won't split into groups. These questions are just for you to consider, you don't need to address them in your write-ups.
- How did the thinking of Rob Pike, Dennis Ritchie, and the rest of the OS researchers at Bell Labs change over time? Specifically, what ideas did they throw away, and what new ideas came to replace them?
- What ideas remained consistent across UNIX, Plan 9, and Inferno?
- Why didn't Plan 9 or Inferno have the success of UNIX?
- What was the impact of Plan 9 and Inferno?
- How does the Styx architecture compare with the architectures of other distributed OSs we've been studying?
- What does Styx ^not^ capture about the design of UNIX, Plan 9, or Inferno?
- Is the view of "everything is a file" still relevant today, or are other approaches (e.g. object-oriented interfaces) more suitable? Why?
Project Topic Discussion
If you are planning on doing a class project, please come to class prepared to give an informal presentation on your chosen topic. If you wish, you may use a formal electronic presentation; if you do so, please either email it to me by Wednesday morning or make it web accessible. Note that you should only plan to speak for about five minutes - so please, no more than five slides!
In your presentation, please address the following issues:
- What is the topic of your project?
- If you are writing a research proposal, what is your idea, and why is it novel?
- Why did you choose it - why do you find it interesting?
- How is it related to distributed operating systems?
- (Very briefly) What related work have you identified?
Everyone will receive a "group participation" grade on Wednesday. Those presenting a project idea will be graded on the basis of their presentation. (Note that slides are not necessary in order to get an A; all you need to do is coherently and concisely present your current thinking on your project topic.)
Everyone else will be graded on the quality of the feedback given to the presenters. I expect everyone not presenting to give at least one constructive comment during class - that will get you a B. More participation, the better the grade - unless you start going overboard. Remember, time will be limited!
Presentations
Presentation slides are here.