Explore the journey of building a unique data engine using Ruby, a graph database, and a bitmap store in this RubyConf 2022 talk. Dive into Zappi's data history, starting with the challenge of disconnected data sets and serialized data frames, and discover how an in-memory index solution was developed. Learn about utilizing RedisGraph to model data relationships and connecting semantically equal nodes. Examine the implementation of a query layer for indexing a bitmap store, resulting in significantly faster dataset interrogation. Follow the evolution from the world before MeasureStore to the introduction of cross-market analysis, harmonization, and the integration of bitmap stores and graphs. Gain insights into the architecture, performance improvements, and future steps for this innovative data indexing approach.
Overview
Syllabus
Intro
Data Indexing with RGB
The world before the MeasureStore
Context
Storage
Introducing MeasureStore
Cross market analysis
Cross comparison
Harmonisation
The bitmap store
The graph
Architecture
Performance
Next steps
Taught by
Confreaks