QuestionQ6

VerifiedDesigning integration solutions to meet performance requirements

A marketing organization is developing a Mule application to handle campaign data. The Mule application periodically checks an SFTP location for a file and processes that file’s records. File size can range from 10 MB to 5 GB. Because vCore availability is limited, the Mule application is deployed to a single CloudHub worker/replica with a 0.2-vCore size.

The application must transform and send different versions of this file to three separate downstream SFTP locations.

What is the most direct way to configure the SFTP operations or event sources to process the large files while supporting these deployment requirements?

Explanation

A file-stored repeatable stream supports multiple reads of the file content for the separate transformations and SFTP deliveries. It keeps only an initial buffer in memory and stores additional content on disk, avoiding the memory pressure that large files would create on a 0.2-vCore worker. MuleSoft documents that repeatable streams support multiple or concurrent reads, and that file-store streaming buffers content to disk once it exceeds the configured in-memory size.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!