Amazon ECR service quotas
The following table provides the default service quotas for Amazon Elastic Container Registry (Amazon ECR).
| Name | Default | Adjustable | Description |
|---|---|---|---|
| Filters per rule in a replication configuration | Each supported Region: 100 | No | The maximum number of filters per rule in a replication configuration. |
| Images per repository | Each supported Region: 10,000 |
Yes |
The maximum number of images per repository. |
| Layer parts | Each supported Region: 4,200 | No | The maximum number of layer parts. This is only applicable if you are using Amazon ECR API actions directly to initiate multipart uploads for image push operations. |
| Lifecycle policy length | Each supported Region: 30,720 | No | The maximum number of characters in a lifecycle policy. |
| Maximum layer part size | Each supported Region: 10 | No | The maximum size (MiB) of a layer part. This is only applicable if you are using Amazon ECR API actions directly to initiate multipart uploads for image push operations. |
| Maximum layer size | Each supported Region: 42,000 | No | The maximum size (MiB) of a layer. |
| Minimum layer part size | Each supported Region: 5 | No | The minimum size (MiB) of a layer part. This is only applicable if you are using Amazon ECR API actions directly to initiate multipart uploads for image push operations. |
| Rate of BatchCheckLayerAvailability requests | Each supported Region: 1,000 per second |
Yes |
The maximum number of BatchCheckLayerAvailability requests that you can make per second in the current Region. When an image is pushed to a repository, each image layer is checked to verify if it has been uploaded before. If it has been uploaded, then the image layer is skipped. |
| Rate of BatchGetImage requests | Each supported Region: 2,000 per second |
Yes |
The maximum number of BatchGetImage requests that you can make per second in the current Region. When an image is pulled, the BatchGetImage API is called once to retrieve the image manifest. If you request a quota increase for this API, review your GetDownloadUrlForLayer usage as well. |
| Rate of CompleteLayerUpload requests | Each supported Region: 100 per second |
Yes |
The maximum number of CompleteLayerUpload requests that you can make per second in the current Region. When an image is pushed, the CompleteLayerUpload API is called once per each new image layer to verify that the upload has completed. |
| Rate of GetAuthorizationToken requests | Each supported Region: 500 per second |
Yes |
The maximum number of GetAuthorizationToken requests that you can make per second in the current Region. |
| Rate of GetDownloadUrlForLayer requests | Each supported Region: 3,000 per second |
Yes |
The maximum number of GetDownloadUrlForLayer requests that you can make per second in the current Region. When an image is pulled, the GetDownloadUrlForLayer API is called once per image layer that is not already cached. If you request a quota increase for this API, review your BatchGetImage usage as well. |
| Rate of InitiateLayerUpload requests | Each supported Region: 100 per second |
Yes |
The maximum number of InitiateLayerUpload requests that you can make per second in the current Region. When an image is pushed, the InitiateLayerUpload API is called once per image layer that has not already been uploaded. Whether or not an image layer has been uploaded is determined by the BatchCheckLayerAvailability API action. |
| Rate of PutImage requests | Each supported Region: 10 per second |
Yes |
The maximum number of PutImage requests that you can make per second in the current Region. When an image is pushed and all new image layers have been uploaded, the PutImage API is called once to create or update the image manifest and the tags associated with the image. |
| Rate of UploadLayerPart requests | Each supported Region: 500 per second |
Yes |
The maximum number of UploadLayerPart requests that you can make per second in the current Region. When an image is pushed, each new image layer is uploaded in parts and the UploadLayerPart API is called once per each new image layer part. |
| Rate of image scans | Each supported Region: 1 | No | The maximum number of image scans per image, per 24 hours. |
| Registered repositories | Each supported Region: 10,000 |
Yes |
The maximum number of repositories that you can create in this account in the current Region. |
| Rules per lifecycle policy | Each supported Region: 50 | No | The maximum number of rules in a lifecycle policy |
| Rules per replication configuration | Each supported Region: 10 | No | The maximum number of rules in a replication configuration. |
| Tags per image | Each supported Region: 1,000 | No | The maximum number of tags per image. |
| Unique destinations across all rules in a replication configuration | Each supported Region: 25 | No | The maximum number of unique destinations across all rules in a replication configuration. |
Managing your Amazon ECR service quotas in the AWS Management Console
Amazon ECR has integrated with Service Quotas, an AWS service that enables you to view and manage your quotas from a central location. For more information, see What Is Service Quotas? in the Service Quotas User Guide.
Service Quotas makes it easy to look up the value of all Amazon ECR service quotas.
To view Amazon ECR service quotas (AWS Management Console)
Open the Service Quotas console at https://console.aws.amazon.com/servicequotas/
. -
In the navigation pane, choose AWS services.
-
From the AWS services list, search for and select Amazon Elastic Container Registry (Amazon ECR).
In the Service quotas list, you can see the service quota name, applied value (if it is available), AWS default quota, and whether the quota value is adjustable.
-
To view additional information about a service quota, such as the description, choose the quota name.
To request a quota increase, see Requesting a quota increase in the Service Quotas User Guide.
Creating a CloudWatch alarm to monitor API usage metrics
Amazon ECR provides CloudWatch usage metrics that correspond to the AWS service quotas for each of the APIs involved with the registry authentication, image push, and image pull actions. In the Service Quotas console, you can visualize your usage on a graph and configure alarms that alert you when your usage approaches a service quota. For more information, see Amazon ECR usage metrics.
Use the following steps to create a CloudWatch alarm based on one of the Amazon ECR API usage metrics.
To create an alarm based on your Amazon ECR usage quotas (AWS Management Console)
Open the Service Quotas console at https://console.aws.amazon.com/servicequotas/
. -
In the navigation pane, choose AWS services.
-
From the AWS services list, search for and select Amazon Elastic Container Registry (Amazon ECR).
-
In the Service quotas list, select the Amazon ECR usage quota you want to create an alarm for.
-
In the Amazon CloudWatch Events alarms section, choose Create.
-
For Alarm threshold, choose the percentage of your applied quota value that you want to set as the alarm value.
-
For Alarm name, enter a name for the alarm and then choose Create.
