[Announcements | Schedule and Readings | Assignments and Quizzes | Syllabus/Policies ]

Reading Questions, Lecture 7 (9/29)

Papers: GiST, R*-Trees

These papers cover several advanced indexing techniques: GiST is a framework that can be used to build a large class of tree-based indices; R*-Trees are a specific tree-based index that are widely used in geo-spatial databases.

Questions to consider:

  1. What are the core methods of the GiST API?
  2. What are the advantages of implementing a tree-based index in GiST rather than building it from scratch? What's hard about building an index data structure that GiST takes care of for you?
  3. What applications are R*-trees useful for?
  4. What performance guarantees can you make about lookup, insert, and delete times in R*-trees? How do those guarantees differ from those of B-trees?

Samuel Madden (madden at csail dot mit dot edu)
Last modified: Mon Sep 27 21:19:08 EDT 2004