I have a pair of systems that do nothing but ingest data and pass on to
others. With the hiccups in data recently, I've spent too much time
looking at pqmon and ldmadmin watch output. Looking at the docs online,
specifically
http://www.unidata.ucar.edu/software/ldm/ldm-6.8.1/basics/monitoring.html#pqmon
it appears that you can use that pqmon output to tune your queue.
The link above states, among other things that "The overhead of managing
the queue could be slightly reduced by recreating the queue with a smaller
number of slots" ... OK, but what does that mean? How many empty slots is
"too many"? I assume that would vary with the size of the queue. Is that
overhead large?
Would I be dramatically improving the service to downstream hosts if I
tweaked this or would I just be nibbling around the edges like Geordi
LaForge adjusting power conversion levels?
--
Peter Laws / N5UWY
National Weather Center / Network Operations Center
University of Oklahoma Information Technology
plaws@xxxxxx
-----------------------------------------------------------------------
Feedback? Contact my director, Craig Cochell, craigc@xxxxxx. Thank you!