AWS Glue

ETL/ELT/Data Transformation

AWS Glue is a serverless extract-transform-load (ETL) service that Amazon Web Services launched in August 2017.

Reviewed by 7wData
Share Email

On this page

Profile

Serverless data extraction, transformation, and loading (ETL) service that discovers, catalogs, and prepares data from 100+ sources for analytics and machine learning.

AWS Glue is a serverless extract-transform-load (ETL) service that Amazon Web Services launched in August 2017. The product has evolved from a pure ETL tool into a broader data-integration platform, now handling discovery, preparation, and pipeline orchestration across on-premises, multi-cloud, and AWS-native data sources. Glue connects to over 100 data sources and stores metadata in its centralized Data Catalog, automating schema detection and code generation to reduce manual ETL work.

By 2026, the service reaches enterprise and mid-market organizations in financial services, healthcare, retail, and media, with named customers including Transport for NSW, Merck, FinAccel, Pagely, and Autodesk. Transport for NSW's 2024 case study reported cutting annual data integration costs by 80 percent after migrating from a third-party ETL platform, achieving faster job deployment and distributed team expertise. Merck uses Glue alongside Kinesis and SageMaker to reduce drug-inspection false rejects by over 50 percent.

In 2025–2026, Glue introduced Zero-ETL integrations that enable near-real-time data replication from operational sources—Aurora, DynamoDB, Salesforce, SAP—directly to Redshift and S3 Tables without manual pipeline code. The service added generative AI assistance through Amazon Q, allowing natural-language job generation and Spark job modernization. AWS Glue 5.1 (November 2025) upgraded the Spark engine to 3.5.6, added Apache Iceberg format-version 3 support with deletion vectors and row-level lineage, and extended Lake Formation fine-grained access control to write operations.

Pricing remains at $0.44 per DPU-hour for standard execution; Flex execution costs $0.29 per DPU-hour (34 percent savings) for batch workloads tolerant of delayed start times. As of 2026, SoftwareReviews gave AWS Glue an 8.7/10 composite rating with 88 percent recommendation likelihood and 99 percent renewal intent. The service faces entrenched competition from Talend (which offers 1,000+ connectors and on-premises deployment) and Databricks (which emphasizes collaborative notebooks and open-table formats). AWS Glue's strength lies in tight AWS service integration and automatic scaling, but adoption outside the AWS ecosystem remains limited.

Track AWS Glue and 240+ vendors.

48k+ subscribers read the daily AI & data note. One email, both newsletters. Unsubscribe anytime.

Products by AWS Glue

Who buys this

  • Enterprise financial institutions consolidating transaction and compliance data from legacy systems into data warehouses
  • Healthcare organizations standardizing patient data from electronic health records, lab systems, and IoT medical devices
  • Government and public-sector agencies automating cost-intensive third-party ETL platform migrations
  • Retail and e-commerce firms unifying online sales, inventory, and customer feedback for demand forecasting
  • Media and entertainment companies cataloging and transforming video, audio, and asset metadata at scale

Publicly disclosed clients

  • Transport for NSW (New South Wales public transport authority)
  • Merck (pharmaceutical manufacturing and drug inspection)
  • FinAccel (fintech, credit scoring)
  • Autodesk
  • Expedia

Strengths and what to watch

Strengths

  • Serverless, pay-per-use architecture with automatic scaling from gigabytes to petabytes, zero infrastructure management overhead
  • Tight integration with AWS ecosystem (Redshift, S3, DynamoDB, Lake Formation, SageMaker) enabling zero-ETL near-real-time data movement and fine-grained access control
  • Generative AI-assisted job creation and Spark modernization via Amazon Q, reducing debugging time from days to minutes

Watch for

  • Learning curve remains steep for teams unfamiliar with Apache Spark, Python, or Scala; deep debugging expertise required for complex pipelines
  • Limited connectivity outside AWS ecosystem; 1,000+ connectors offered by competitors like Talend reduce lock-in risk for multi-cloud deployments
  • AWS Glue for Ray sunsetting April 30, 2026; no new customer sign-ups accepted, signaling strategic de-emphasis of that execution engine

Recent moves

Key Information

Industry
ETL/ELT/Data Transformation
Founded
2017
Employees
224906
Headquarters
Seattle, Washington
Country
United States

Frequently Asked Questions

What is AWS Glue?

AWS Glue is a serverless extract-transform-load service that discovers, catalogs, and prepares data from 100+ sources for analytics and machine learning. It automates schema detection and code generation, eliminating manual ETL work while connecting to on-premises, multi-cloud, and AWS-native data sources.

What is zero-ETL in AWS Glue?

Zero-ETL integrations enable near-real-time data replication directly from operational sources like Aurora, DynamoDB, and Salesforce to Redshift and S3 Tables without requiring manual pipeline code. This approach eliminates intermediate processing steps and significantly reduces latency for analytics and machine learning workloads.

How much does AWS Glue cost?

AWS Glue charges $0.44 per data processing unit hour for standard execution. Flex execution, designed for batch workloads tolerant of delayed start times, costs $0.29 per DPU-hour, delivering 34 percent savings. Costs scale automatically with data volume from gigabytes to petabytes.

Can AWS Glue use artificial intelligence?

Yes, AWS Glue integrates Amazon Q, a generative AI assistant that enables natural-language ETL job creation and automated Spark job modernization. This reduces debugging time from days to minutes and eliminates manual code optimization, making job creation faster for users unfamiliar with Apache Spark.

Does AWS Glue work outside AWS?

AWS Glue connectivity outside the AWS ecosystem is limited. The service excels at integrating AWS services but offers fewer connectors than competitors like Talend. For multi-cloud deployments requiring broad third-party source connections, users often face lock-in constraints or choose alternative ETL platforms.

How did Transport for NSW benefit from AWS Glue?

Transport for NSW migrated from a third-party ETL platform to AWS Glue and cut annual data integration costs by 80 percent. The migration also accelerated job deployment from months to days and enabled distributed team expertise, demonstrating Glue's advantage for cost-heavy legacy platform replacements.

How AWS Glue compares

Direct head-to-head against 3 competitors. Picked by 7wData.

This company

AWS Glue

Positioning
Serverless data extraction, transformation, and loading (ETL) service that discovers, catalogs, and prepares data from 100+ sources for analytics and machine learning.
Customer segments
Enterprise financial institutions consolidating transaction and compliance data from legacy systems into data warehouses
Strengths
Serverless, pay-per-use architecture with automatic scaling from gigabytes to petabytes, zero infrastructure management overhead
Watch for
Learning curve remains steep for teams unfamiliar with Apache Spark, Python, or Scala; deep debugging expertise required for complex pipelines
Recent moves
AWS Glue zero-ETL integrations now available in Asia Pacific (Mumbai) region, expanding real-time data movement for analytics and ML workloads

Databricks

Positioning
Unified lakehouse platform offering ETL, SQL analytics, and ML in one environment, targeting enterprise data engineering teams.
Customer segments
Enterprise data and ML teams at financial services, healthcare, and technology firms running Apache Spark workloads.
Strengths
Lakeflow declarative pipelines for batch and streaming ETL, paired with Unity Catalog for cross-workspace governance.
Watch for
Compute costs on large workloads outpace comparable ETL services; customers report billing surprises at petabyte scale.
Recent moves
Completed $5B Series L equity raise at $134B valuation, February 2026, funding Lakebase serverless database development.

Fivetran

Positioning
Automated ELT platform built around managed connectors and schema drift handling for cloud data warehouse loading.
Customer segments
Mid-market and enterprise data teams needing fast connector deployment without building or maintaining custom ingestion pipelines.
Strengths
500-plus pre-built connectors with automated schema change handling; minimal engineering overhead for ongoing pipeline maintenance.
Watch for
Row-based pricing escalates sharply at high data volumes; customers with large syncs report unexpected cost overruns.
Recent moves
Announced merger with dbt Labs, October 2025, combining ELT ingestion and SQL transformation under one vendor.

Qlik Talend Cloud

Positioning
Cloud-agnostic ETL and data quality platform, bundled with Qlik analytics after the 2023 acquisition.
Customer segments
Healthcare and financial services enterprises requiring data quality certification, lineage, and compliance audit trails alongside ETL.
Strengths
1,000-plus connectors with native data quality rules and on-premises-to-cloud hybrid deployment, unlike most cloud-native ETL services.
Watch for
Post-acquisition roadmap uncertainty and product overlap; Talend Open Studio free edition discontinued, disrupting open-source-dependent teams.
Recent moves
Qlik Talend Cloud reached general availability at Qlik Connect, May 2025, unifying ETL and analytics under one product.

Sources

  1. aws.amazon.com — AWS Glue product overview, capabilities, and features
  2. aws.amazon.com — Named customers (FinAccel, Merck, Pagely, Autodesk, Expedia) with testimonials and case study references
  3. www.softwarereviews.com — 2026 customer reviews, 8.7/10 composite rating, 88% recommendation, 99% renewal intent, emotional footprint scores
  4. aws.amazon.com — Transport for NSW 80% cost reduction, migration from third-party ETL platform, deployment acceleration from months to days
  5. aws.amazon.com — AWS Glue launch date (August 14, 2017), initial pricing ($0.44/DPU-hour), serverless architecture, schema detection and code generation
  6. aws.amazon.com — REST API connector announcement February 2026, 60+ native connectors for 100+ data sources
  7. aws.amazon.com — Amazon Q integration for natural-language ETL job generation, Spark job modernization, AI troubleshooting
  8. aws.amazon.com — AWS Glue 5.1 Apache Spark 3.5.6 upgrade, Apache Iceberg format version 3, Hudi 1.0.2, Delta Lake 3.3.2 support