The combination of storage/retrieval/calculation, executed close to the actual data, is what gives 20x advantage. CPU cache is hard-to-believe fast, see: http

The combination of storage/retrieval/calculation, executed close to the actual data, is what gives 20x advantage. CPU cache is hard-to-believe fast, see: https://stratoflow.com/latency-numbers-every-java-developer-should-know/ https://stratoflow.com/latency-numbers-every-java-developer-...