Question when testing Caching Server with resperf

[Eivind Olsen]

> run with query 100-thousand -> maximum throughput ~ 9000  -> named
process ~ 450 MB
> run with query 100-thousand -> "ran out query data" errors
> run with query 3-millions -> maximum throughput ~ 9000 -> named process
~ 400 MB
> run with query 3-millions -> maximum throughput ~ 16000 -> named process
~ 500 MB

First, I know it's not what you asked for, but why only run with
100-thousand queries in the input? If I remember correctly, resperf builds
up gradually in speed towards 100.000 queries per second, unless you
override that speed-limit.

Now, on to what might be your issue. What kind of data do you query for?
Is it real external data, or data you ensure is already in the cache? Are
all the queries unique, or are there duplicates meaning the 1st query for
a specific name will have an empty cache and need to go external and the
2nd query for the same name will get a faster answer from the cache?
Did you flush the cache between tests?

Regards
Eivind Olsen