Amazon Glue concepts - Amazon Glue
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

Amazon Glue concepts

The following diagram shows the architecture of an Amazon Glue environment.

The basic concepts populating your Data Catalog and processing ETL dataflow in Amazon Glue.

You define jobs in Amazon Glue to accomplish the work that's required to extract, transform, and load (ETL) data from a data source to a data target. You typically perform the following actions:

  • For data store sources, you define a crawler to populate your Amazon Glue Data Catalog with metadata table definitions. You point your crawler at a data store, and the crawler creates table definitions in the Data Catalog. For streaming sources, you manually define Data Catalog tables and specify data stream properties.

    In addition to table definitions, the Amazon Glue Data Catalog contains other metadata that is required to define ETL jobs. You use this metadata when you define a job to transform your data.

  • Amazon Glue can generate a script to transform your data. Or, you can provide the script in the Amazon Glue console or API.

  • You can run your job on demand, or you can set it up to start when a specified trigger occurs. The trigger can be a time-based schedule or an event.

    When your job runs, a script extracts data from your data source, transforms the data, and loads it to your data target. The script runs in an Apache Spark environment in Amazon Glue.


Tables and databases in Amazon Glue are objects in the Amazon Glue Data Catalog. They contain metadata; they don't contain data from a data store.

Text-based data, such as CSVs, must be encoded in UTF-8 for Amazon Glue to process it successfully. For more information, see UTF-8 in Wikipedia.

Amazon Glue terminology

Amazon Glue relies on the interaction of several components to create and manage your extract, transform, and load (ETL) workflow.

Amazon Glue Data Catalog

The persistent metadata store in Amazon Glue. It contains table definitions, job definitions, and other control information to manage your Amazon Glue environment. Each Amazon account has one Amazon Glue Data Catalog per region.


Determines the schema of your data. Amazon Glue provides classifiers for common file types, such as CSV, JSON, AVRO, XML, and others. It also provides classifiers for common relational database management systems using a JDBC connection. You can write your own classifier by using a grok pattern or by specifying a row tag in an XML document.


A Data Catalog object that contains the properties that are required to connect to a particular data store.


A program that connects to a data store (source or target), progresses through a prioritized list of classifiers to determine the schema for your data, and then creates metadata tables in the Amazon Glue Data Catalog.


A set of associated Data Catalog table definitions organized into a logical group.

Data store, data source, data target

A data store is a repository for persistently storing your data. Examples include Amazon S3 buckets and relational databases. A data source is a data store that is used as input to a process or transform. A data target is a data store that a process or transform writes to.

Development endpoint

An environment that you can use to develop and test your Amazon Glue ETL scripts.

Dynamic Frame

A distributed table that supports nested data such as structures and arrays. Each record is self-describing, designed for schema flexibility with semi-structured data. Each record contains both data and the schema that describes that data. You can use both dynamic frames and Apache Spark DataFrames in your ETL scripts, and convert between them. Dynamic frames provide a set of advanced transformations for data cleaning and ETL.


The business logic that is required to perform ETL work. It is composed of a transformation script, data sources, and data targets. Job runs are initiated by triggers that can be scheduled or triggered by events.

Job performance dashboard

Amazon Glue provides a comprehensive run dashboard for your ETL jobs. The dashboard displays information about job runs from a specific time frame.

Notebook interface

An enhanced notebook experience with one-click setup for easy job authoring and data exploration. The notebook and connections are configured automatically for you. You can use the notebook interface based on Jupyter Notebook to interactively develop, debug, and deploy scripts and workflows using Amazon Glue serverless Apache Spark ETL infrastructure. You can also perform ad-hoc queries, data analysis, and visualization (for example, tables and graphs) in the notebook environment.


Code that extracts data from sources, transforms it, and loads it into targets. Amazon Glue generates PySpark or Scala scripts.


The metadata definition that represents your data. Whether your data is in an Amazon Simple Storage Service (Amazon S3) file, an Amazon Relational Database Service (Amazon RDS) table, or another set of data, a table defines the schema of your data. A table in the Amazon Glue Data Catalog consists of the names of columns, data type definitions, partition information, and other metadata about a base dataset. The schema of your data is represented in your Amazon Glue table definition. The actual data remains in its original data store, whether it be in a file or a relational database table. Amazon Glue catalogs your files and relational database tables in the Amazon Glue Data Catalog. They are used as sources and targets when you create an ETL job.


The code logic that is used to manipulate your data into a different format.


Initiates an ETL job. Triggers can be defined based on a scheduled time or an event.

Visual job editor

The visual job editor is a graphical interface that makes it easy to create, run, and monitor extract, transform, and load (ETL) jobs in Amazon Glue. You can visually compose data transformation workflows, seamlessly run them on Amazon Glue's Apache Spark-based serverless ETL engine, and inspect the schema and data results in each step of the job.


With Amazon Glue, you only pay for the time your ETL job takes to run. There are no resources to manage, no upfront costs, and you are not charged for startup or shutdown time. You are charged an hourly rate based on the number of Data Processing Units (or DPUs) used to run your ETL job. A single Data Processing Unit (DPU) is also referred to as a worker. Amazon Glue comes with three worker types to help you select the configuration that meets your job latency and cost requirements. Workers come in Standard, G.1X, G.2X, and G.025X configurations.