Have fun with in-memory columnar databases or SQL engines and see how fast they are (the ones that use CPU-efficient data structures and data element access, pl

Have fun with in-memory columnar databases or SQL engines and see how fast they are (the ones that use CPU-efficient data structures and data element access, plus SIMD processing). For example Apache Arrow datafusion [1] Edit: Also, run a cluster of anything (in VMs or containers) and muck around killing individual cluster nodes or just suspending them/throttling them to be extremely slow to simulate a brownout/straggling cluster node. [1] https://github.com/apache/arrow-datafusion https://github.com/apache/arrow-datafusion