Using the AWS SDK for Java

The AWS SDK for Java provides an API for the Amazon S3 bucket and object operations. For object operations, in addition to providing the API to upload objects in a single operation, the SDK provides API to upload large objects in parts (see Uploading Objects Using Multipart Upload API). The API gives you the option of using a high-level or low-level API.

Low-Level API

The Low-level APIs correspond to the underlying Amazon S3 REST operations, such as create, update, and delete operations that apply to buckets and objects. When you upload large objects using the low-level multipart upload API, it provides greater control such as letting you pause and resume multipart uploads, vary part sizes during the upload, or to begin uploads when you do not know the size of the data in advance. If you do not have these requirements, use the high-level API to upload objects.

High-Level API

For uploading objects, the SDK provides a higher level of abstraction by providing the TransferManager class. The high-level API is a simpler API, where in just a few lines of code you can upload files and streams to Amazon S3. You should use this API to upload data unless you need to control the upload as described in the preceding Low-Level API section.

For smaller data size the TransferManager API uploads data in a single operation. However, the TransferManager switches to using the multipart upload API when data size reaches certain threshold. When possible, the TransferManager uses multiple threads to concurrently upload the parts. If a part upload fails, the API retries the failed part upload up to three times. However, these are configurable options using the TransferManagerConfiguration class.