Building a 100‑Petabyte Analytics Warehouse

on YouTube ↗49m

" summary

Criteo describes building one of Europe’s largest Hadoop clusters with Hive as the central data-processing tool, explaining how their Hive platform evolved from fragile installations on dedicated servers into a best-in-class architecture. The talk covers making the system self-healing and adding automatic scaling to handle growing workloads at petabyte scale.

AI-written summary · about

" chapters (5)

1. 0:00Introduction
2. 10:50Analytics platform evolution
3. 21:17Hive architecture challenges
4. 34:19Testing Hive versions
5. 42:21Q&A

" related talks

• 57mBig Data: Threat or Opportunity?Shaun Connolly · PASS Business Analytics Conference 2013
• 45mHadoop in the Cloud: The Story of Migrating Hundreds of PetabytesMikhail Maryufich · HighLoad++ 2023
• 59mHadoop Essentials for the SQL Server ProfessionalBill Preachuk · 24 Hours of PASS: Summit Preview 2015
• 33mAnalytics Platform at Petabyte ScaleSharad Agarwal · Cypher 2017
• 07mIn-Hadoop Analytics: Bringing analytics to big dataAnjul Bhambhri · Strata Conference 2014
• 15mThe Future of DataDoug Cutting · Strata Conference in London 2013
• 45mBuilding an Analytics PlatformJeff Klukas · PyOhio 2016
• 45mSolving Real & Big (Data) Problems Using HadoopEva Andreasson · GOTO Aarhus 2013