Assignment #3: Project Status

Due Date: 4/7/2005 (or earlier)

Signup Sheet


At this point, you should be making good progress on your assignment. In particular, whatever infrastructure you need, whether a real prototype, or a simulator, etc., should be in place. Also, you should have found and read related work - at least three papers, preferably more. You should be thinking about how you are going to evaluate your project - what kinds of experiments will you be running towards the end? Finally, any initial results that you have will quite useful to you, to generate more feedback. Realize, the more you have done, the better feedback you will get from me.

In this assignment, you will have to sign up for a face-to-face meeting with me during the week of 4/4 (up until 4/7) and discuss the points mentioned above. Come prepared, as this counts as part of your overall project grade.

When: Wednesday or Thursday (use the Signup Sheet ). Expect to meet for 10-15 minutes.



The first step towards a successful project is a sound infrastructure within which to work. Make sure to have everything that you need in place by this time -- preferably, this will be finished long before the status meeting.

Note: By infrastructure, I mean a solid base towards the final goal. For example, if you are going to study caching policies via a simulation, it would be good to have the simulation environment in place and working. Or if you are building a new interface to a file system, it would be good to have studied the kernel and have a good plan as to how to proceed, or to even have made some initial progress on modifying the system. This does not mean the entire implementation of whatever it is you are doing!

Related Work

Related work is an important part of any research effort. Find and study as many papers as you can that are related to what you are doing. Take notes, and even start sketching out how you would write a "related work" section in your final paper.

A good related work section is more than a laundry list - you should really try to "synthesize" and "categorize". That is, you should be try to put together other people's work in your own viewpoint, and in a more general frame of thought. Remember to emphasize what has been done in the past, and more importantly what hasn't. Look through some of the papers we have read in class for good examples.


Finally, all good projects will have a good empirical evaluation to them. It is not just enough to build "it" and assume "it" will work. You have to measure it! For this assignment, start writing down ideas on how you would measure what you are doing, and be ready to discuss them.

Initial Results

If you have really gotten going (like you should have), you would have some initial results. First graphs from a prototype system or simulator, or measurements of an existing system to justify what you are doing. The more results you have, the more feedback you will get from me on what you should be doing (and what you should avoid).