Cost-Effective Updating of Distributed Reordered Indexes

Explore cost-effective strategies for updating distributed reordered indexes in this 22-minute conference talk. Delve into index reordering techniques that optimize document collection numbering, enhancing inverted index compression. Examine the challenges of maintaining effective reorderings as collections grow over time, particularly in distributed retrieval systems. Learn about methods for preserving and reinstating reorderings, backed by experimental results from a large English news article corpus. Gain insights into the impact of reordering on query execution time and consider various update operations, including batch append. Discover practical approaches to balance index efficiency and maintenance costs in evolving document collections.

Intro
Inverted Indexing
Document Reordering
Distributed Retrieval Systems
Update Operations
Questions to consider
Data and Experiments
Batch Append Operations
Plus, One More Thing