COMP-614: Distributed Data Management

Project Guidelines


Outline

The project work is an essential part of this course. Its aim is to get you started with doing independent research / work in the area of distributed systems and distributed information systems. During the first part of the project you will do some literature research within a specific topic and write a survey report about your readings (around 25 pages, 12pt, 1.5 spacing). You will present the survey to the class in form of a lecture with discussion. During the second part of the project you should elaborate a new problem and its solution within this research area.  There are three alternatives:
  1. A research proposal for a relatively complex problem. I do not expect a complete solution but a description of the approach you would take to obtain the solution. Your work should demonstrate an understanding of the research area and an insight into the problem such that given enough time (2 to 5 more months), you could carry it to its logical conclusion and complete the research.  The delivery is a report.
  2. A complete solution to a small-sized, more specialized problem. The delivery is a report.
  3. An implementation and evaluation of algorithms presented in some research paper. The delivery is a report with an overview of the implementation and the evaluation results.
The resulting report should also contain around 20-25 pages.

Schedule

Talk Schedule, see main page

On Pursuing Research

Literature Search


Below you will find several possible research topics from which you can pick one. For each of the topics I will later provide at least three research papers that can serve as starting points for your literature study. I expect you to include at least one more paper in your research review. If you find other papers related to the topic, that you think are better, feel free to change. For your talk and your basis for your own study, feel free to look at more papers or even choose different papers.  You should look at the proceedings of the following conferences. For databases oriented topics ICDE, VLDB, SIGMOD ... for more distributed system topics Middleware, ICDCS, DSN, SOSP. You do not need to look at journals but if you see an interesting sounding journal paper, the following journals are good: IEEE and ACM Transactions on.... journals (transactions on database systems, transactions on computer systems, etc.),  Information Systems, IEEE TKDE, VLDB Journal. A good starting point for you own literature search is google scholar or  Michael Ley's Computer Science Bibliography . The first one is maintained automatically, and hence, also contains a lot of duplicate information, etc. The latter is more structured and consistent. Citeseer is also a good starting point. You will find many papers online. You will also have access to many of the research papers through the McGill library system. A more recent resource is
McGill has subscribed to the digital libraries for ACM, IEEE, and Springer. You can go directly to their webpages and have access if you connect from a computer within McGill

Research Topics