Re: PATCH: pgbench - random sampling of transaction written into log

On 03.09.2012 01:40, Tomas Vondra wrote:
> So, here is a fixed version of the patch. I've made these changes:

Committed with some minor kibitzing.

> 1) The option is now '--sampling-rate' instead of '-R' and accepts float
> arguments. I've decided not to use 0.01 = 1% but use 1 = 1%, so it
> accepts values between 0 and 100. I find this more natural.

I find a fraction, ie. 0.01 = 1% more natural, so I changed it back to
that. I realize this is bike-shedding, but I believe Robert also found
this more natural, as that's what he suggested upthread.

I edited the comments and variable names a little bit, and also added a
small paragraph to the pgbench user manual about this under
"Per-Transaction Logging", in addition to the explanation under
"Benchmarking Options"

- Heikki