We’re excited to share the complete materials from our session at Flo 2025: Data’s Grand Tour — Navigating the Modern Data Architecture Maze. This 45-minute presentation covers the full landscape of modern data architectures, from foundational concepts to production decision frameworks.
🎤 Session Details
Aakash Khandelwal & Jyotika Agarwal
19th September 2025
45 Minutes
Flo 2025
🚀 Presentation Resources
🖥️ Live Presentation
Navigate through the slides below using the arrow keys or the on-screen controls. Press F for full screen.
📋 Session Agenda
This session was structured as a journey through the modern data architecture landscape:
| # | Section | Topics Covered |
|---|---|---|
| 1 | Data Jargon 101 | Schema-on-Read/Write, ETL vs ELT, OLTP vs OLAP, ACID, Medallion Architecture |
| 2 | 6 Core Architectures | Data Lake, Modern Data Warehouse, Lakehouse, Data Mesh, Data Fabric, Microsoft Fabric |
| 3 | Comparative Analysis | Feature matrix across all architectures — cost, governance, performance, ML fit |
| 4 | Decision Framework | How to choose: use case mapping, team maturity, cloud ecosystem alignment |
| 5 | Table Formats Revolution | Delta Lake, Apache Iceberg, Apache Hudi — internals, trade-offs, and when to use each |
🗺️ The Architecture Landscape
The session covered these six core architectures and how they relate to each other:
🏞️ Data Lake
Store everything in raw format at low cost. Schema applied at query time. Foundation for ML and exploratory analytics.
🏢 Modern Data Warehouse
Schema-on-write governance. MPP query execution. Optimised for structured BI and reporting workloads.
🏠 Data Lakehouse
ACID transactions on open formats (Delta/Iceberg). Unified platform for SQL, ML, and streaming on a single data copy.
🕸️ Data Mesh
Decentralised domain ownership. Data as a product. Self-serve platform. Federated governance at scale.
🧩 Microsoft Fabric
Unified SaaS analytics on OneLake. Data Engineering, Warehouse, Real-Time Analytics, Science, and Power BI in one platform.
📋 Table Formats
Delta Lake, Iceberg, Hudi — open formats that bring ACID transactions, time travel, and schema evolution to the data lake.
📚 Go Deeper: Concept Articles
Each architecture from this session has been expanded into a full deep-dive concept article, available in the Concepts section under the Data Architecture series:
Vocabulary & Fundamentals 02 · Data Lake
Zones, Tools & Trade-offs 03 · Modern Data Warehouse
MPP, Dimensional Modelling 04 · Data Lakehouse
Open Table Formats & ACID 05 · Data Mesh
Domain Ownership & Products 06 · Microsoft Fabric
OneLake & Unified Platform 07 · Table Formats
Delta, Iceberg & Hudi Internals 08 · Decision Framework
Comparison & How to Choose 09 · Architecture Patterns
Medallion, Lambda, Kappa, Z-Order 10 · Real-Time Architecture
Kafka, Flink & Streaming Patterns 11 · Unstructured Data & Vectors
Embeddings, RAG & Vector DBs 12 · Business Glossary
Plain-English Definitions & Examples 13 · Learning Path
Roadmap & Curated Resources