Project

Requirement

A main component of the course is a research-oriented course project. You will work as a team of 2-3 students on a topic related to operating systems or systems in general. The project should identify a clear problem, motivate why that problem is important and why existing solutions are insufficient, propose an idea to address the problem, implement a prototype, and demonstrate the idea’s efficacy through experiments and comparisons with existing solutions. In the end, you will write up a project report about your work, just like the research papers you read in the class.

Please use the USENIX conference paper template to format your project proposal and report.

Expectations

The best outcome for a project is inventing the next Unix (seminal, high-impact work)! A more realistic great outcome is that the project will be suitable for publication at a major systems conference (OSDI, SOSP, EuroSys, NSDI, ASPLOS, USENIX ATC). Indeed, some papers in top systems conference are a result of a course project (e.g., Infiniswap and Siloz were projects started in EECS 582 in the past). In those cases, additional work after the course is often needed to push the project to be more complete for publication, but the course project should lay the groundwork.

The minimum expectations for the project are: (1) it should be interesting! (2) it should produce a software artifact. In other words, even if the project does not yield a real publication, it should have intellectual depth and something that the readers can learn from. For (2), the resulting software artifact should be usable.

A survey of existing literature will not be suitable for the course project. Measurement studies of systems or system techniques are allowed with instructor’s approval. In that case, you must still build a measurement tool as part of the project.

Topics and Tips

I will provide a list of suggested topics, but you are encouraged to come up with your own!

  Tips
  • When reading the papers in this course, try to identify the (sub-)problems are related but unsolved, limitations of the proposed solution, a possible better way to address the problem, whether the paper's insights or techniques are applicable in a different problem domain. Thinking along those lines can give you inspirations about potential project topics.
  • Besides reading papers, your own experiences can also be a source to brainstorm about the project topics. For example, you encountered annoying performance issues in system software and were frustrated that the existing profilers could not pinpoint the root causes; so a potential project is to identify the limitations in existing profilers and propose a solution to address them.
  • Form the project team as soon as possible and start the project early. Systems projects take time, even for a course project. If you start late, it is likely that you will not be able to finish the project in time.
  • While you are encouraged to be ambitious, make sure the project is manageable, such that you have something to present at the end.

I will meet with each team to help refine the project scope. I will always be available for project discussions.

Deliverable #1: Proposal

In the proposal, briefly describe the following elements:

  • the problem you plan to tackle, and why the problem is important
  • the state of the art for solving this problem, and why the current solution is insufficient
  • the approach/idea you propose
  • the objectives/expected outcomes
  • the potential challenges
  • the timeline and milestones
  • division of work among the team members
  • a list of related work that you plan to read

The proposal should be around 2 pages and include your group member’s name and email. The proposal only needs to be submitted by one of the group members.

Deliverable #2: Checkpoint 1 Report

Submit a checkpoint report (around 3 pages) to describe the status of your project, the solution you have build so far, some initial results, new challenges that arise, any adjustments to your approach.

Deliverable #3: Checkpoint 2 Report

The second checkpoint report is closer to the final report. It should include all the progress you have made so far (including content from the first checkpoint report), and highlight the new results since the previous checkpoint, and the remaining tasks and timeline for completing the project.

Deliverable #4: Final Report

The final report should look like a research paper. It motivates the importance of the problem with concrete evidence or data points. It explains why existing solutions do not solve the problem well. Then it describes the insights and ideas of your work, the challenges you need to address, how did you address them, why your idea is better, what are your system design and implementation. Then shows the experiment results, interpret the results, describe the observations and implications, conclude whether or how the results demonstrate your solutions’ superiority over state-of-the-art solution. Lastly, include a related work section to compare and contrast with your solution.

There are no hard restrictions on the page limit on the final report, because papers do not need to be long to be good! But your final report must comprehensively cover all the points mentioned above. The rule of thumbs is that when it does so, the report is around 6 to 10 pages.

Deliverable #5: Presentation/Poster

Each project team will be given a slot to do a presentation in class to showcase your ideas and results to others. A live demo of the system you developed would be a nice element of the presentation. We might host a poster session in Tishman Hall if conditions allow.

Deliverable #6: Source Code

You will be required to turn in your source code along with a detailed README and scripts for grading. The software artifact should be runnable by the TA and instructor.