Hi everyone, I'm Hesanda Liyanage, an undergraduate student, and I've recently started contributing to Apache James with my first pull request on the project for JAMES-4231, focusing on S3 object compaction.
The current work introduces S3 object compaction by packing small standalone blobs into larger chunk objects, with virtual slot addressing and ranged reads. I've also been working on the associated Cassandra reference updates, crash-safety/coherence handling, GC recompaction, integration tests, and documentation. PR: https://github.com/apache/james-project/pull/3193 I'm continuing to work through the review feedback and improve the implementation, particularly around correctness testing and property-based testing. Looking further ahead, Benoit suggested that this work could potentially evolve into a Google Summer of Code project for 2027, with the longer term goal of exploring S3 GC/compaction at larger scale using Apache Spark. I'd be very interested in continuing this work with the James community and learning more about the project's GSoC process and expectations as we get closer to the 2027 cycle. Looking forward to getting to know everyone and contributing more to James. Best, Hesanda
