[rabbitmq-discuss] RabbitMQ experience

[Martin Sustrik]

> On another tack entirely, when measuring toxicity, the only sensible way
> of specifying how poisonous something is is to use LD50, the "lethal
> dose 50%", i.e. that dose at which there is a 50% chance of death (or at
> which 50% of your dosed population dies, etc.). (Also used are LD75,
> LD90 etc.)
> 
> Is there some kind of LD50-like concept that could be used in measuring
> peak capacity?

Hm, interesting. I've never thought of that kind of metric. Possibly 
something like "if there's a peak lasting N seconds containing M 
messages there's a 50% chance that the max latency will exceed X 
microseconds". The obvious problem is that there are to many free 
variables in the formula.

Fixing X to say 1 millisecond makes sense (customers tend to have quite 
a clear idea what's the maximal viable latency), however, I'm not sure 
how to combine N and M into a single scalar value describing quality of 
the peak.

Martin