When transferring data across a wide area network in particular, it is much more efficient to compress the data prior to sending it across the network. Decompression can then be done at the destination.
These components allow for two things:
1. Extracting data directly into a compressed file directly from an SSIS pipeline, and extracting data from these compressed files directly into an SSIS pipeline. This eliminates the need for a second step to cab or compress the files after extraction or to uncab or uncompress prior to loading. It also places the metadata in the compressed file so it can be read by the compressed file source.
2. Because a sequential access file is used, it allows for multiple BLOB columns to be included in the compressed file.
The components use the .NET System.IO.Packaging.Package library to create, write to, and read from the compressed file. The data is written as a binary stream (to minimize space required for numeric data) and compressed using a deflate compression stream. The metadata needed for the SSIS pipeline is stored in XML format in a separate package part from the data.
These components are especially useful for two stage data movement into, and out of Azure.