Amazon Machine Learning endpoints and quotas - AWS General Reference

Amazon Machine Learning endpoints and quotas

The following are the service endpoints and service quotas for this service. To connect programmatically to an AWS service, you use an endpoint. In addition to the standard AWS endpoints, some AWS services offer FIPS endpoints in selected Regions. For more information, see AWS service endpoints. Service quotas, also referred to as limits, are the maximum number of service resources or operations for your AWS account. For more information, see AWS service quotas.

Service Endpoints

Region Name Region Endpoint Protocol
US East (N. Virginia) us-east-1 machinelearning.us-east-1.amazonaws.com HTTPS
Europe (Ireland) eu-west-1 machinelearning.eu-west-1.amazonaws.com HTTPS

Service Quotas

Resource Default
Data file size* 100 GB
Batch prediction input size 1 TB
Batch prediction input (number of records) 100 million
Number of variables in a data file (schema) 1,000
Recipe complexity (number of processed output variables) 10,000
Transactions Per Second for each real-time prediction endpoint 200
Total Transactions Per Second for all real-time prediction endpoints 10,000
Total RAM for all real-time prediction endpoints 10 GB
Number of simultaneous jobs 25
Longest run time for any job 7 days
Number of classes for multiclass ML models 100
ML model size 2 GB

The size of your data files is restricted to ensure that jobs finish in a timely manner. Jobs that have been running for more than seven days are automatically terminated, resulting in a FAILED status.

For more information, see Amazon ML Quotas in the Amazon Machine Learning Developer Guide.