so, i've been playing around with PERCENT in graphs and started thinking 
about another possibility for 95th percentile calculations.  i wonder 
how possible it would be to have an RRA store 95th percentile instead of 
average.  this would make rolled-up data, like a 1-day 95th percentile 
value from a days worth of 5-min samples much more interesting in the 
long run than average.  especially when talking about interface traffic 
levels, IMHO.  possibility?  thoughts?

+m

--
Unsubscribe mailto:[EMAIL PROTECTED]
Help        mailto:[EMAIL PROTECTED]
Archive     http://www.ee.ethz.ch/~slist/rrd-developers
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi

Reply via email to