By default, Apache Cassandra implements LZ4 compression with a chunk_length_in_kb size of 64. It's important to note that in read-heavy or mixed workloads, using a smaller chunk size can help to improve (reduce) our overall disk I/O usage. This is especially true if your data model is performing small, frequent reads. Jon Haddad of The Last Pickle put it best when he wrote the following:
Ultimately, we need to ensure we use a chunk length that allows us to minimize our I/O. Larger chunks can compress better, giving us a smaller disk footprint, but we end up needing more hardware, so the space savings becomes meaningless for certain workloads. There’s no perfect setting that we can apply to every workload. Frequently, the most reads ...
