Skip to content
datastudyguides

DP-750 · 4.1.2 · Deploy and maintain data pipelines and workloads

Notebook or pipeline

Choose between notebook and Lakeflow Spark Declarative Pipelines

Use notebooks for custom imperative workflows and Lakeflow pipelines for declarative batch or streaming data transformations with managed operations.

See the declarative experience

Microsoft Learn Lakeflow Pipelines Editor showing source files, pipeline graph, execution controls, and Genie Code.

Lakeflow Spark Declarative Pipelines center the workflow on maintained datasets and their dependencies rather than on manually sequenced execution steps.

Choose the programming model

  • Notebook

    Style
    Imperative
    Best for
    Custom procedural logic
    Dependencies
    Explicit orchestration
    Quality
    Custom code

  • Lakeflow pipeline

    Style
    Declarative
    Best for
    Maintained data graph
    Dependencies
    Inferred from datasets
    Quality
    Native expectations
Notebook or declarative pipeline

Use notebooks when custom APIs, branching code, specialized libraries, or non-tabular control are central.

Avoid manual orchestration when the graph already knows

If one maintained dataset reads another, let the declarative dependency graph determine execution order.

Key points

  • Notebook: Flexible imperative code and custom control
  • Lakeflow pipeline: Declarative dataset dependencies
  • Pipeline features: Managed updates, quality, lineage, monitoring
  • Choice: Prefer pipeline when dataflow semantics fit

Exam trap

Do not wrap every transformation in orchestration notebooks when declarative datasets can express the dependency graph directly.

Sources · 2 Microsoft Learn pages Show ▾

Checked on October 4, 2026. Report a problem

Go deeper on Microsoft Learn

Checked against Microsoft Learn on October 4, 2026. Report a problem with this card

How well do you know this?