Yes, that is correct. As you noted, Aurora performs quite well on TPC-C due to its quorum-based write acknowledgment and on TPC-H when data modifications are minimal. However, CH workloads introduce contention, write amplification, and read/write reconciliation under analytical scan pressure, which is a difficult use-case for Aurora to handle.

That said, in our tests we only saw 2x improvements in CH benchmarks. However, we found out that it was due to an architectural issue in our VM I/O path and how we virtualize the storage. Based on our estimations we should see ~5x difference but for that we need to revamp our storage virtualization first.

We have plans for publishing a CH benchmark results on a follow up blog post. However, we didn't want to do that for now to not put misleading results.