A6 » A6TEAM_AWESOME

A6TEAM_AWESOME

Last modified by Brionna Lopez on 2010/10/06 09:27

A6TEAM_AWESOME

To-Do

  1. study the document “Projects for the HCC Foundation Course” (in “Students’ Projects” on the Wiki)
  2. select a project that is of real interest to you
  3. find collaborators and form a group of peers (such a group may already exist) with whom you will work on the project
  4. sign up for the project (if it is still available) on the site “Students’ Projects
  5. identify parts of the project and an initial role distribution between your team members
  6. each team member should contribute
    1. a brief rational what she/he finds interesting
    2. what she/he plans to do to get started on the project
Which project did you select?
Project-1: Warehouses with Interactive Content
Describe the parts of the project you have identified and the initial role distribution among your team members
Main areas to explore: system architecture (ie how does netflix distribute streaming content?), usability (how to make large amounts of content accessible? netflix queue, youtube suggestions, search...), dynamic elements (users upload videos, rate movies etc), how well do companies accomplish their goals?
For each team member, provide a brief rational what she/he finds interesting
Ian: There are many aspects of this project that are interesting to me. First off, presenting such an enormous amount of content to the user in an efficient way is a huge challenge and I want to study the best practices of doing so. I would also like to investigate the architecture of warehouse-style systems and see how companies manage to get such large amounts of data distributed.
Aaron Vimont: The idea of a data warehouse is still rather new to me. I have heard the term before but have not looked into how they work.  I find that most interesting.  I use sites like YouTube and Netflix, but I never looked into how they are structured or how they provide data so efficiently. I am especially intrigued as to how the rating and suggestions work. Working with a group will should also be good because we can discuss how everything works to gain a better understanding of data warehouses. We will have an opportunity to see how data warehouses will change in the future. I may never work on a data warehouse project again, but having the opportunity to learn how they work will no doubt be beneficial to me in the future.

Adam Jackson: This project interests me in the sense that, while I widely use both services, I have never examined either from a computer science point of view. It will be interesting to consider how they both go about accomplishing what it is they need to do from a technical aspect. Taking it a step further and analyzing whether or not they are successful based on their operational models should also yield interesting results.

For each team member, provide a statement of what she/he plans to do to get started on the project
Ian: I'd like to get started by looking in general at accessibility of large warehouses of content, in particular things like the Netflix suggestion algorithm, YouTube suggestions etc.
Aaron Vimont: I will get started by researching the dynamic aspects of data warehouses.  Making a data warehouse dynamic can add a lot of complication to the setup.  Updates become more difficult, as well as the changes to the schema.  I will try to gain a solid understanding of how people have handled these problems and work that can be done in the future.

Adam Jackson: I will get started on this project by researching the physical aspects of YouTube's and Netflix's data warehouses. I will ascertain where and how many physical data centers they operate as well as what it means for a single consumer to access data that may be spread across multiple locations. From there, I will try to determine the relative "success" of each business in terms of availability, speed, and ease of use for end consumers.

Tags:
Created by Ian Smith on 2010/10/04 16:19

This wiki is licensed under a Creative Commons 2.0 license
XWiki Enterprise 2.7.1.${buildNumber} - Documentation