Skip to content
DataEngr.com

DataEngr.com

A working reference for data engineers. 200 terms, each with a definition you can use in a meeting and an explanation you can build from.

Pipelines, storage, modeling, orchestration, governance, and the lakehouse and AI architectures they are all converging on. Written and maintained by Alex Merced, co-author of Apache Iceberg: The Definitive Guide.

Start Here

The terms most people look up first. Each one links out to the concepts around it, so you can follow a thread as far as you need to.

Common Questions

What does a data engineer actually do?

A data engineer builds and operates the systems that move data from the places it is created into the places it gets analyzed. That covers ingestion, storage, transformation and serving, plus the orchestration, testing and monitoring that keep those pipelines producing correct results on every run, not just the first one.

Read the full entry

What is the difference between ETL and ELT?

ETL transforms data on a separate tier before loading it, so the destination only ever stores modeled tables. ELT loads raw data into the warehouse or lakehouse first and transforms it there using the platform's own compute. ELT keeps a replayable raw copy; ETL discards it at transform time.

Read the full entry

What is a data lakehouse?

A data lakehouse is open files on object storage with a table format layered over them. The table format, usually Apache Iceberg, tracks every data file in metadata rather than inferring a table from a directory, which restores ACID transactions, schema evolution and time travel to data that any engine can still read.

Read the full entry

Do I need streaming, or is batch enough?

Batch is enough whenever the business can act on data that is minutes or hours old, and it is cheaper and simpler to operate. Streaming earns its complexity when a decision has to happen within seconds of the event. Micro-batching sits between the two and covers a large share of real requirements.

Read the full entry

Where should I start if I am new to data engineering?

Start with how analytical data is shaped and where it lives: data modeling, the difference between a warehouse, a lake and a lakehouse, and how a pipeline is scheduled. Those three ideas explain most architecture decisions you will meet, and every other term connects back to them.

Read the full entry

Worth Reading

Semantic Layer: The Definitive Guide

Why a shared metric layer sits above the tables rather than inside each BI tool.

Read Article

Apache Polaris: The Catalog Standard

How an open catalog keeps the lakehouse readable by more than one engine.

Read Article

Table Formats Explained

What a table format does, and the problems it was invented to fix.

Read Article

What is Dremio?

A look at the unified lakehouse platform and where it fits in a stack.

Read Article

What Apache Iceberg Native Actually Means

Separating real Iceberg support from a connector with an Iceberg label on it.

Read Article

Open Source and The Lakehouse

How open standards changed who gets to read your data.

Read Article

What is Agentic Analytics?

What changes when an agent, not a person, is the one asking the questions.

Read Article

Communities & Events

Join the Conversation

Upcoming Events