The relational database in big data projects is still the system of record, and most analytics platforms copy their core business data from it. This post shows what a relational database does well, where it sits in a modern platform, and how the SQL Server 2025 json type adds a flexible schema next to ordinary tables.
What Is NoSQL? Data Models, the CAP Theorem and Trade-Offs
What Is NoSQL? It is a family of data models other than the classic table, each built for one kind of access pattern. This post covers the four data models, eventual consistency and the CAP theorem in plain words, and how to choose.
ETL vs ELT: How Big Data Changed the Way We Load Data
ETL vs ELT comes down to when you transform: before the load on a separate engine, or after it inside the platform. This post compares the two and runs a small ELT pipeline in T-SQL on JSON orders.
Storing JSON in SQL Server
Storing JSON in SQL Server can use validated nvarchar or the native 2025 json type, with indexed paths and relational columns for stable facts.
JSON_ARRAYAGG and JSON_OBJECTAGG: Building JSON From Rows in SQL Server 2025
Build customer arrays with JSON_ARRAYAGG and settings objects with JSON_OBJECTAGG in SQL Server 2025, with explicit order and null rules.
Change Logs as JSON: Old and New Values in One Column
Skip the audit table with a column for every field. One JSON payload per changed row is lighter, if you handle NULL and multirow updates correctly.
Structured, Semi-Structured and Unstructured Data Explained
Structured, Semi-Structured and Unstructured Data differ in where the schema lives and how you query them. This post compares the three shapes and runs one query across a table, a json column and free text in SQL Server 2025.