|
Project
Project Preliminary Ideas: Due 9AM Tuesday, October 15th (or earlier) -- 5% of Project Grade
Proposal: Due 9AM on Tuesday, October 22nd (or earlier) -- 10% of Project Grade
Progress Report: Due 9AM on Tuesday, November 12th -- 10% of Project Grade
Lightning Talk: Due 9AM Tuesday, December 3rd -- 5% of Project Grade
Talks: Tuesday, December 10th, 9AM - 2PM -- 30% of Project Grade
Final Project Report: Due 5pm Tuesday, December 17th -- 40% of Project Grade
Weight: 35%
Project Goal
The project has two options. The first option is to do some original
research in computer architecture (e.g.,
write a simulator and gather some simple numbers) or to re-validate data
found in a published paper.
This type of project is especially recommended for students
considering a Ph.D.
You will be graded on how well you define
your problem, survey previous work, design and conduct experiments, and
present your results.
The goal to shoot for is a conference paper, like the ones
in your reader. Since time is limited, however, the above goal
is hard to reach, and I will reward those that aim high even if
they do not completely succeed.
The key is insuring that some aspects of your work are completely
done; it is very hard to grade a project where the simulator
did not quite work.
This type of project should use groups of two.
With my permission, project groups of one or three are permissible.
In the second option, you may work as a group of two
to write a 15-20 page paper
that surveys an area within computer architecture.
The paper should:
summarize work in an area, giving extensive references to support your arguments,
present opinions of others for and against various options
(with references), and
conclude with your opinion of the strengths and weaknesses of
arguments presented above.
In the second option, you may work a group of two to write a paper
that surveys an area within computer architecture. The paper should:
summarize work in an area, giving extensive references, present opinions
of others for and against various options (with references), and conclude
with your opinion of the strengths and weaknesses of arguments presented above.
You will be graded on the completeness of your survey, the accuracy of your summaries,
the support you give for your opinions, and the quality of your presentation. Since a
survey paper is "safer" than a research project, I will hold survey papers to a higher
standard of completeness and analysis of the literature.
For this project, you may turn in your progress reports and final written report up to
24 hours late with a 15% penalty. I strongly recommend that you do not avail yourself
of this option; your hard work deserves full credit.
Preliminary Project Ideas
Send your project ideas by email, use subject: [CS758 Project Ideas].
You should prepare 2-3 project ideas that you would like to pursue. It is fine to have a
preferred topic, but you must think about at least one alternative. You should also tell me
the names and email addresses of each of the project team members.
Project Proposal
Send your proposal by email, use subject: [CS758 Project Proposal] Lastname1/Lastname2.
(If there are more than two project participants, please include their names as well.)
Proposals should be about two pages long (10 point font, single spaced, 1 inch margins, either single or double-column per page, any of the standard formats like IEEE is fine). They should include:
-
Names and email addresses of each of the project team members,
-
A description of your topic,
-
A statement of why you think the topic is interesting or important,
-
For research projects, a description of the methods you will use to evaluate your ideas
-
References to at least three relevant papers you have already read
-
A plan to address other related work.
Accompanying your proposal submission, you should also review the proposal checklist to ensure that your proposal contains all of the above components. You should submit the following checklist rubric with your proposal.
The synthesis lecture and the papers we've read point to many other papers that you may find useful.
See also Proceedings of the International Symposium on Computer Architecture,
Proceedings of the High-Performance Computer Architecture, Proceedings of the conference on
Architectural Support for Programming Languages and Operating Systems, IEEE Transactions on Computers,
ACM Transactions on Computer Systems, IEEE Computer, IEEE Micro, IEEE Transactions on Parallel and
Distributed Systems. Near the end of the course web page are a couple pointers to on-line resources
that may also be useful.
Project Progress Report
Submit your Progress Report on Canvas
Progress Reports should include a revised version of the proposal plus one page describing
accomplishments so far and any unresolved problems that you have encountered. Concentrate on
describing sub-tasks completed, rather than the tasks started. For example, say "completed simulator
design at subroutine level" rather than saying "started writing simulator".
It is also common for the list of related work to have expanded to 5-8 papers.
Your progress report will be graded using the following rubric:
Progress Report Grading Rubric
Project Lightning Talk
Upload your Lightning Talk to Canvas
Computer Architecture conferences are increasingly utilizing online lightning talks to motivate
your conference talk. To help you prepare for this, your course project will also have a
lightning talk. Your lightning talk should be a 90 second version of your talk designed to
encourage people to attend your talk. The best talks will provide a memorable introduction
to your talk. As incentive for making a good lightning talk, the best lightning talk
will get to choose when they present during the Project Talks.
The only rules for the Lightning Talk are: a) it has to be 90 seconds or less, b) it cannot contain any inappropriate language, remarks, etc. and c) it should be done in a format that could potentially be uploaded to YouTube (since that's what conferences are doing nowadays). Beyond these requirements, you can be as creative as you would like. The goal is to stand out, while also relating the lightning talk to your project -- if your lightning talk is completely unrelated to the project, it likely won't help encourage people to attend. If only one person from the group appears in the video, then please send me an email (similar format to the other emails) explaining what the other project member(s) role(s) was.
Your lightning talk will be graded using the following rubric:
Lightning Talk Grading Rubric (Note: this rubric may be updated)
Some examples of recent lightning talks, both good and bad, can be found here: ISCA 2018 and MICRO 2018.
Before, when the talks were done as part of the conference, I gave several lightning talks, which you can view here:
Once I get permission, I will also post some lightning talks from groups in 752 last Fall.
Project Talks
We will schedule a mini-conference of 20-minute conference-style talks. Of this time, 16 minutes should be spent presenting and 4 minutes will be spent answering questions.
All group members should deliver part of the talk and answer some of the questions. The talk should give highlights of the final
report, including the problem, motivation, results, conclusions, and possible future work. Time limits
will be enforced to let everyone present. Please, practice your talk to make it better and see how long
it is. Have a plan for what slides to skip if you get behind.
We will be using the same presentation grading rubric as prior semesters:
Presentation Grading Rubric
Upload the final slides 90 minutes before the presentation to Canvas. Not uploading your slides to Canvas by this time will reduce your project presentation grade by 10%. Additionally, please also bring a printout of your slides to the presentations so I can write comments on them as you present. Preferably 4 slides per page. It does not need to be in color, grayscale or black and white is fine.
Please see Oral
Presentation Advice, including David Patterson's How to Give a
Bad Talk and K. Compton and M. L. Chang's
Terrible
Presentations (...and how to not give one).
A rough timeline for the presentations is below (all times are approximate, but the order between groups is fixed). The presenting groups will be filled in based on people's selections -- recall that those with the best lightning talks get to choose when they want to present.
| Time Slot |
Group |
| 900 AM - 920 AM |
Setup
|
| 920 AM - 940 AM |
Preyesh & Rohan |
| 940 AM - 10 AM |
Devin & Giri |
| 10 AM - 1020 AM |
Kevin & Shivangi |
| 1020 AM - 1040 AM |
Overflow |
| 1040 AM - 11 AM |
Gaurav & Kyle M. |
| 11 AM - 1120 AM |
Rahul & Haocheng |
| 1120 AM - 1140 AM |
Nick & Martin |
| 1140 AM - 12 PM |
Overflow |
Final Project Report
Upload your PDF (and the checklist) to Canvas
Reports should consist of an abstract, body and optional appendices, much like a
conference paper.
The abstract should summarize the contributions of the report in
one or two paragraphs. The length of the body should be the equivalent of 15-20 pages
at double-spaced 10-point. The body must include a section that describes which tasks each group member worked on. If multiple group members worked on the same task, please explain what each group member did for that task. Additional supporting material of any length can be put in
appendices. I will read the body and only skim appendices.
Your final report will be graded using the following rubric:
Final Report Grading Rubric.
You should submit the final report checklist with your final report, in order to ensure that your report contains all the expected components.
If you would like help improving your writing, I encourage you to take advantage of the
UW Writing Center. Professor Reps also has a good set of
reference slides
on how to write papers, which you may want to look at.
The report must be written in your own words. Please review the Writing Center guidelines
on appropriate quotation, paraphrasing, and citation of prior work
(PDF).
Failure to substantially follow these guidelines will be treated as academic misconduct.
Topics
You are encouraged to come up with your own topic. Ideally, the topic will be related
to your current research interests.
There can be value in combining your project with other work in one of two ways.
First, if you have active research in a related area, you can do a
deeper project by combining your current (supported) research with your course project.
Second, it is also possible to do a single multi-disciplinary project to satisfy multiple
courses (e.g., CS736). However, this must be coordinated with the other instructor(s) and
you will be expected to do two (or more) projects worth of work. In all such cases, it is
imperative that such overlaps be disclosed at the proposal stage so that we can address the
scope of research that will be required.
|
