Basic Information
| Field | Details |
|---|---|
| English name | AWS Glue |
| Full name | AWS Glue |
| Chinese description | 无服务器数据集成与 ETL |
| Japanese description | サーバーレス ETL |
| Exam frequency | ⭐⭐⭐⭐ |
| Often confused with | EMR / Data Pipeline |
In one sentence
A serverless data integration service: Data Catalog describes data, while Glue ETL cleans and transforms it.
Key points
- Crawlers inspect sources and infer schemas; Data Catalog stores metadata such as location, format, columns, and types.
- Glue ETL extracts, transforms, and loads data for Athena, Redshift, or machine learning.
- The Catalog does not store primary business data; the actual data remains in S3 or another source.
Exam focus
- Metadata, schema, and data catalog point to Glue Data Catalog.
- Serverless ETL and transformation point to Glue; compare EMR when full Spark or Hadoop control is needed.
Common pitfalls
- Do not choose by product name alone; confirm data type, latency, control, operations, and cost constraints.
Remember
Catalog is the directory; ETL is the processing plant.