Multi-Region Access Point request routing - Amazon Simple Storage Service

Multi-Region Access Point request routing

When you make a request through a Multi-Region Access Point, Amazon S3 determines which of the buckets that are associated with the Multi-Region Access Point is closest to you. Amazon S3 then directs the request to that bucket, regardless of the AWS Region it is located in.

After the Multi-Region Access Point routes the request to the closest-proximity bucket, Amazon S3 processes the request as if you made it directly to that bucket. Multi-Region Access Points aren't aware of the data contents of an Amazon S3 bucket. Therefore, the bucket that gets the request might not contain the requested data. To create consistent datasets in the Amazon S3 buckets that are associated with a Multi-Region Access Point, you can configure S3 Cross-Region Replication (CRR). Then any bucket can fulfill the request successfully.

Amazon S3 directs Multi-Region Access Point requests according to the following rules:

  • Amazon S3 optimizes requests to be fulfilled according to proximity. It looks at the buckets supported by the Multi-Region Access Point and relays the request to the bucket that has the closest proximity.

  • If the request specifies an existing resource (for example, GetObject), Amazon S3 does not consider the name of the object when fulfilling the request. This means that even if an object exists in one bucket in the Multi-Region Access Point, your request can be routed to a bucket that doesn't contain the object. This situation will result in a 404 error message being returned to the client.

    To avoid 404 errors, we recommend that you configure S3 Cross-Region Replication (CRR) for your buckets. Replication helps resolve the potential issue when the object that you want is in a bucket in the Multi-Region Access Point, but it's not located in the specific bucket that your request was routed to. For more information about configuring replication, see Configuring replication for use with Multi-Region Access Points.

    To ensure that your requests are fulfilled by using the specific objects that you want, we also recommend that you turn on bucket versioning and include version IDs in your requests. This approach helps ensure that you have the correct version of the object that you are looking for. Versioning-enabled buckets can also help you recover objects from accidental overwrite. For more information, see Using S3 Versioning in S3 buckets.

  • If the request is to create a resource (for example, PutObject or CreateMultipartUpload), Amazon S3 fulfills the request by using the closest-proximity bucket. For example, consider a video company that wants to support video uploads from anywhere in the world. When a user makes a PUT request to the Multi-Region Access Point, the object is put into the bucket with the closest proximity. To then make that uploaded video available to others around the world for download with the lowest latency, you can use CRR with bidirectional (two-way) replication. Using CRR with two-way replication keeps the contents of all the buckets that are associated with the Multi-Region Access Point synchronized. For more information about using replication with Multi-Region Access Points, see Configuring replication for use with Multi-Region Access Points.