[ltt-dev] performance results I measured with the latest lttng patches

[Jan Blunck]

Sorry for the late reply.

On Wed, Sep 24, Michael Rubin wrote:

> I am curious what the expectations and goals the lttng efforts have
> for performance in these scenarios. Do you guys have a goal for perf
> impact in the situations of not configured (or not compiled) vs
> configured (compiled in) but no trace points enabled vs tracepoints
> enabled?

The important thing is that we don't have performance regressions when having
tracepoints configured but no probes connected. If we have a performance
impact on this one, I'll not be able to defend the patches for default
enablement.

> What sort of benchmarks are you using to test performance? That way we
> can use the same ones to better compare notes. Do you have access to
> HW with lots of cores?

At the moment I only have a 64 core single image system available to test
with. I don't have time right now but already blocked some time for beginning
of next week.

Regards,
	Jan

-- 
Jan Blunck <jblunck at suse.de>