Data Warehousing and Business Analysis · Data Mining

Metadata Reporting

Watch lectures

Metadata Reporting

Metadata is data about data. In a warehouse it describes the source, meaning, structure, transformation, ownership, quality and usage of the stored data.

Important

Be able to define metadata, classify it into technical, business and operational types with examples, and explain the role of the metadata repository and data lineage.

NoteExample

For the column FACT_SALES.amount the metadata records: source ERP.SALES_LINES.NET_AMT, type NUMBER(12,2), currency INR, rule qty * unit_price * (1 - discount), refresh daily at 02:00, owner Finance, quality rule amount >= 0.

1. Types of Metadata

Type Also called Answers Contents
Technical Structural, system How is data stored and moved? Table and column names, datatypes, keys, indexes, ETL mappings, partitions
Business Semantic What does it mean? Definitions, KPIs, hierarchies, owners, calculation rules, glossary
Operational Administrative Did it run? Job status, run times, rows read, written and rejected, error logs
Process / lineage Data lineage Where did it come from? Source-to-target mappings, transformation rules, dependencies, versions

2. Metadata Repository

MermaidDiagram
Rendering diagram…

3. Why Metadata Reporting Matters

Lineage and impact analysis

If a source column is dropped, exactly which ETL jobs, cubes and reports will break?

4. A Typical Metadata / Lineage Report

Table Type Source Transformation Refresh Rows loaded Rows rejected DQ % Owner
DIM_CUSTOMER Dim (SCD-2) CRM.CUST_MASTER Trim, gender standardize, surrogate key Daily 02:00 12,540 7 99.94 Sales Ops
FACT_SALES Fact ERP.SALES_LINES + POS files Join, currency convert, derive amount Daily 02:30 480,213 132 99.97 Finance
DQ%=RacceptedRtotal×100%,Lag=tnow−tlastLoad,SLA met  ⟺  Lag≤SLADQ\% = \frac{R_{accepted}}{R_{total}} \times 100\%, \qquad Lag = t_{now} - t_{lastLoad}, \qquad SLA\ met \iff Lag \le SLA

5. Lineage Graph

MermaidDiagram
Rendering diagram…

6. Standards and Tools

Standard / tool Purpose
CWM (Common Warehouse Metamodel, OMG) Standard metamodel for exchanging warehouse metadata
XMI XML format used to exchange CWM models
RDF / OWL Semantic metadata, ontologies, business glossary
Informatica Metadata Manager, IBM Metadata Workbench ETL and BI metadata, lineage
Collibra, Alation, DataHub, Apache Atlas Data catalog and governance platforms
Exam TrapTypical questions
  • Define metadata; classify it into technical, business and operational types with examples (5-7 marks).
  • Explain the role of the metadata repository in ETL and query tools (5 marks).
  • Short note on data lineage and impact analysis (3-5 marks).

Frequently Asked Questions

QWhat is metadata?

Data about data: it describes the source, structure, meaning, transformation, quality and ownership of warehouse data.

QWhat are the types of metadata?

Technical (structure and mappings), business (definitions and rules), operational (job runs and errors) and process or lineage metadata (source-to-target flow).

QWhat is data lineage?

A record of where data came from and how it was transformed on its way to a report or table.

QWhat is a metadata repository?

A central store that keeps all metadata so ETL tools, the warehouse and BI tools can share and use it.

QName a metadata standard.

CWM (Common Warehouse Metamodel) by the OMG, exchanged using XMI.

QHow does metadata help query tools?

The tool reads business names, joins and hierarchies from metadata, so users can build queries without knowing the physical schema.

AI LearningWeb Platform

Coursify — Personal AI Course Generator

Struggling with a difficult subject or concept? Turn any engineering topic into a structured course with interactive lessons, quizzes, and research-backed explanations.

← All Data Mining modules