First, pre-tuning numbers for the *small* "velocibunny"

Ok, I know you have been asking …
What is a “velocibunny”
Think of it as … um … a very very fast JackRabbit. Not that JackRabbit isn’t fast … it appears to be best in its class in performance.
But “velocibunny” is faster.
A lot faster.

It is specifically designed to be usable as a very fast database engine. As in PostgreSQL and related.
Now mind you, it is in lab with the first iteration of an OS load, and
NO TUNING
This is “build the RAID10 out of the box and start using it with a base load”.
But, how fast is it?
Random reads, 8k size across 128 GB of storage space (machine has 16 GB, RAID cards have much smaller cache … that is, these results are not cached)
Using a simple fio script

[random]
size=8g
directory=/data
iodepth=768
ioengine=sync
direct=1
blocksize=8k
numjobs=16
nrfiles=1
group_reporting


We measure 38,582 IOPs.
We do expect 46k or so after tuning. Also, we expect to be able to measure for the larger unit soon as well.
Think ~90k IOPs if we have done our homework right.