search for: statustics

Folks I've started to use "vdo" instead of zfs in Centos 7. I hope this is a wise decision. However, I'm a bit mystified in decoding the "vdostats" output. I'd like to figure out how well deduplication is working. One measure would be to find two numbers: L = How many bocks are in use as reported to tools like df P = How many actual blocks are in use.

I have a data set that has student test scores along with several categorical variables. I would like to generate a set of summary stats (mean, variance, n) for the data grouped by school authority and by exam topic. I have tried the by() function but that seems to only be able to handle one level of grouping. In particular what I would like is something like the following Board Subject Mean