Data-intensive Computing Systems: Project
| Course information |
| Course schedule and notes |
| Assignments |
| Readings |
| Project |
| Extra Materials |
Project 1
This project is a semester-long project where students are expected to identify and address an important research problem. The project will involve a major programming effort. A system demonstration as well as a research report is due at the end of the course. Projects that get high grades will, possibly with some more effort, lead to a research publication at a major conference.The project milestones are:
- A project proposal (with possibly a short presentation) due by October 10.
- A mid-project update to be sent to the instructor in the first week of November.
- A system demonstration as well as a research report due soon after classes end. The deadline will be announced later.
Students will work in teams comprising two or three students each.
Project 2
This project is a one-month project done by each student individually. (Group work is not allowed even though it is possible that multiple students may be assigned the same topic.) Each student will be assigned one topic that is proposed as an improvement to MapReduce and Hadoop. The student will do an in-depth study on this topic and create a very informative web-site on the topic. While the format and contents of the web-site are left to the student's creativity and effort, the web-site should at least contain the following material:
- Important uses/applications of this topic.
- All important architectural, design, and algorithmic aspects of this topic. How do these aspects help improve performance, fault tolerance, ease-of-use, etc., over MapReduce and Hadoop?
- Some experimental results that you generated about this topic by doing experiments on the Amazon cloud.
- A comprehensive list of technical references (research papers, blog posts, JIRA entries, etc.) about the topic. The references should be ranked in decreasing order of how useful you found each one to be.