Relevant Data Analysis: Apache Solr Analytics #bbuzz

on YouTube ↗33m

" summary

Search engine–native analytics inside Apache Solr 7 implements a MapReduce-like pipeline that maps per-document fields, reduces across shards, and merges results to produce Solr Analytics Component aggregations at Bloomberg production scale. Associative reducers (sum, count, min, max) aggregate near-linearly with number of shards; non‑associative reducers such as median and percentile require sorted lists or heavy merging and can use shared reduction data to cut network transfer. Supports value/range/query/pivot facets, multivalued fields, variable functions, and first‑order analytics over hundreds of millions of docs.

AI-written summary · about

" chapters (6)

1. 0:00Intro
2. 4:31SOLR ANALYTICS
3. 15:08DISTRIBUTED ANALYTICS
4. 20:17PERFORMANCE CONSIDERATIONS
5. 23:53ADDITIONAL FEATURES
6. 27:40USE CASES

" related talks

• 29mBuilding Analytics Applications (...)Amrit Sarkar · Berlin Buzzwords 2019
• 20mGain speed and space with NLP in Solr #bbuzzTobias Kässmann · Berlin Buzzwords 2016
• 31mScaling Facets to the Stars 🌟Shikhar Srivastava · Berlin Buzzwords 2021
• 53mVisualizing the Output of Your Search AIMarcus Eagan · Berlin Buzzwords 2019
• 38mAdd ALL the things: abstract algebra meets analyticsAvi Bryant · Strange Loop 2013
• 45mProfessional Data: Using Elasticsearch's Aggregation FrameworkMark Harwood · GOTO Amsterdam 2015
• 27mTales at Scale: Analytics at 1000 QPS and BeyondGian Merlino · Open Source Summit Latin America 2022
• 42mMake Sense of your (BIG) data!David Pilato · Codemotion Roma 2014