Terminal based took for monitoring data through a pipeline:
port:fred ~/code/c/stuffc> ./yes | pv > /dev/null ^C.1GiB 0:00:15 [3.16GiB/s] [ <=> ] port:fred ~/code/c/stuffc> /usr/bin/yes | pv > /dev/null ^C35MiB 0:00:12 [44.8MiB/s] [ <=> ]
Was using pv to test through put with a yes.c based on reddit thread talking about the speed of yes.c. For me an interesting aspect of this is creating a buffer so that you can stuff many more characters through the process. This allow's GNU yes to be far faster, but the code is also larger 77loc compared to13loc.
Is the current speed of yes actually problematic?
$Id: diary,v 1.26 2016/11/14 23:40:17 fred Exp $