I was losing something when because I was reading the line of the
GZIPInputStream outside of the busy while loop. I changed it and now I
am having more throughput. It is also a good idea to use VisualVM to
check if the throughput is correct and where I am losing more cycles.
while
Hi Felipe,
the file is just 80 MBs. It is probably cached in the linux page cache,
there should not be any disk IO involved.
So you are saying is that you can not further increase the throughput for
sleeps shorter than 2000 nanoseconds.
Have you tried running this w/o any Sleep / nano.time
Hi,
I am trying to benchmark a stream application in Flink. So, I am using
the source Function that reads events from the NYC Taxi Rides
(http://training.ververica.com/trainingData/nycTaxiRides.gz) and I
control the emission with System.nanoTime(). I am not using
Thread.sleep because Java does