I just came across this slide about drizzle where they claim to achieve
sub-millisecond latency and they compare with Flink,
https://www.slideshare.net/SparkSummit/drizzlelow-latency-execution-for-apache-spark-spark-summit-east-talk-by-shivaram-venkataramanThe normal drizzle still performs a bit worse than Flink, but the
optimized-drizzle goes really crazy with the latency.
How is this even possible? I thought that such performance is almost not
possible because of the interaction of data over the network and the time it
takes.
Any insights and whether Flink has any plans to respond back?
Thanks & Regards
Biplob
--
Sent from:
http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/