The tie ins with the google dataflow paper are at least superficial (if not deeper on deeper reading) in that they mention: "latency, correctness and cost" as t

The tie ins with the google dataflow paper are at least superficial (if not deeper on deeper reading) in that they mention: "latency, correctness and cost" as their drivers: http://blog.acolyer.org/2015/08/18/the-dataflow-model-a-practical-approach-to-balancing-correctness-latency-and-cost-in-massive-scale-unbounded-out-of-order-data-processing/ http://blog.acolyer.org/2015/08/18/the-dataflow-model-a-prac...