What is Amazon EMR used for?

Amazon EMR is a managed big data platform on AWS for running Apache Spark, Hive and Presto jobs, interactive SQL queries and machine learning workloads at scale. It is a good fit for running Spark jobs on a data lake, moving on-premises Hadoop workloads to AWS and interactive SQL over large datasets.

How much does Amazon EMR cost?

Amazon EMR is a paid product with no free plan. Usage-based. EMR on EC2 adds $0.105 per hour for a c4.2xlarge instance on top of EC2 costs. EMR Serverless costs $0.052624 per vCPU-hour and $0.0057785 per GB-hour.

Is Amazon EMR open source?

No. Amazon EMR is proprietary (closed-source) software and can't be self-hosted. Open-source alternatives to Amazon EMR include Apache Spark and Apache Hive.

What are some alternatives to Amazon EMR?

Amazon EMR competes with Databricks, Cloudera and Google Cloud Dataflow. For open-source options, see Enlisted's ranked list of open-source Amazon EMR alternatives.

Amazon EMR: features, pricing and alternatives

About Amazon EMR

Amazon EMR is a cloud big data platform from AWS that runs open-source analytics frameworks for large-scale distributed processing. Its name comes from Elastic MapReduce, and it descends from the Hadoop ecosystem, but today it is used mainly with frameworks such as Apache Spark, Apache Hive and Presto.

AWS positions EMR for three kinds of work: large distributed data processing jobs, interactive SQL queries and machine learning applications. Because it is a managed service, AWS handles setting up the cluster software, while the team chooses which frameworks to use and how much capacity to run. It commonly reads from and writes to data stored in Amazon S3.

EMR is a proprietary managed service available only on AWS, though the frameworks it runs are open source. Pricing is usage-based under AWS pricing, with the vendor publishing a dedicated EMR pricing page.

Key features

Managed Apache Spark and Hive clusters
Presto for interactive SQL queries
Large-scale distributed data processing
Machine learning workload support
Integration with Amazon S3 data lakes

Good fit for

→Running Spark jobs on a data lake
→Moving on-premises Hadoop workloads to AWS
→Interactive SQL over large datasets

Tags: aws; big-data; spark; hadoop; hive; presto; data-processing; cloud-service

Amazon EMR: questions and answers

What is Amazon EMR used for?: Amazon EMR is a managed big data platform on AWS for running Apache Spark, Hive and Presto jobs, interactive SQL queries and machine learning workloads at scale. It is a good fit for running Spark jobs on a data lake, moving on-premises Hadoop workloads to AWS and interactive SQL over large datasets.
How much does Amazon EMR cost?: Amazon EMR is a paid product with no free plan. Usage-based. EMR on EC2 adds $0.105 per hour for a c4.2xlarge instance on top of EC2 costs. EMR Serverless costs $0.052624 per vCPU-hour and $0.0057785 per GB-hour.
Is Amazon EMR open source?: No. Amazon EMR is proprietary (closed-source) software and can't be self-hosted. Open-source alternatives to Amazon EMR include Apache Spark and Apache Hive.
What are some alternatives to Amazon EMR?: Amazon EMR competes with Databricks, Cloudera and Google Cloud Dataflow. For open-source options, see Enlisted's ranked list of open-source Amazon EMR alternatives.