Redis benchmark needed

[GoogleCodeExporter]

I think we really need a benchmark script for Redis able to measure both
latency percentiles and queries/seconds with different number of clients.
There are at least two important reasons for having it: without a good
benchmark it is very hard to try a coding effort with the goal of improving
Redis performances now or in the future, and it is very hard to tell from a
release to another one if changed has some kind of impact on performances.

This need arise in this days because I'm reworking a lot the Redis
internals in order to use everywhere refcounted objects. This will be a
very important step in order to make the code more coherent and well
designed and to support zero-copy master <-> slave communication channel
when replication will be implemented. From what I can tell sometimes to
have keys in Redis objects is a win, but some other time just an overhead.
A benchmark can tell what's the reality!

Original issue reported on code.google.com by anti...@gmail.com on 3 Mar 2009 at 10:47

[GoogleCodeExporter]

Just a note: probably this stuff should be written in C in order to make the
latencies that can be result of the dynamic programming language implementation 
a non
issue. That is with C we can exactly control the syscalls we use for the 
benchmark,
use static allocation, in order to make as much as possible the only variables 
in
play the ones that happen server-side.

I can write it, just I need the help of some people with experience in latency 
and
this kind of issues. Probably Erlang guys are good at this.

Original comment by anti...@gmail.com on 3 Mar 2009 at 10:52

[GoogleCodeExporter]

Original comment by anti...@gmail.com on 12 Mar 2009 at 9:15

• Changed state: Verified