How eMAG Accelerates Time To Insight & Leaves No Opportunities Uncovered With Dremio
summary
eMAG replaces heavy Qlik reports with Dremio as a SQL semantic layer, joining hundreds of on‑prem and cloud sources (Hadoop, S3, MySQL, SQL Server, Salesforce) into virtual datasets and materialised Reflections. The platform queries over one petabyte in Hadoop, supports ~150 daily users running 30,000 queries, cuts time‑to‑insight from months or weeks to hours, and saves roughly 100 man‑days/month by eliminating manual exports; teams deploy lightweight proofs of concept in a day and isolate heavy workloads while maintaining hybrid on‑prem/cloud flexibility.
AI-written summary · about
chapters (18)
- 0:00Introduction and eMAG overview
- 2:35Legacy BI challenges at eMAG
- 5:53Team structure and bottlenecks
- 7:21The 4P overview problem
- 9:52Requirements for a new solution
- 12:00Testing query engines and choosing Dremio
- 14:28Current architecture with Dremio
- 15:55Example: Call center data integration
- 17:43Reflections and virtual datasets
- 18:51Transparent pipeline improvements
- 19:13Results: Self-service analytics
- 20:59Time to insight improvements
- 22:48Data analyst empowerment
- 23:30Key metrics and savings
- 24:33Reverse Conway maneuver impact
- 25:16Future plans and hybrid infrastructure
- 26:01Operational best practices
- 27:06Closing and Q&A
related talks
13mBuild your open data lakehouse with Dremio and AWS (PRT085)
21mPutting big data self-service analytics in the hands of business people
48mHow smava boosts business agility and innovation w/Amazon Redshift
05mDriving the On-Demand Economy with Predictive Analytics
21mOptimizing data delivery for faster, better analytical insights
16mTurn that Data into a Differentiator - How Analytics Can Drive Product Value
32mConnecting Silos In Real-time With Data Virtualization
35mConquering All Stores with Gimel – A Unified Data Processing Platform