Re: Tracing Ceph results

[Mohamad Gebai <mgebai@suse.com>]

On 05/02/2017 08:49 PM, Mark Nelson wrote:
> On 05/02/2017 04:05 PM, Mohamad Gebai wrote:
>>
>> I've done some benchmarking of LTTng in the past. On average, a kernel
>> tracepoint costs less than 100ns and a userspace one is around 150ns. It
>> also hits the disk to periodically flush the tracing buffers. The
>> overhead is definitely non negligeable. I'll write a how-to as there are
>> quite a few steps to get all of this running.
>
> Awesome job Mohamad!  It'd be sweet if we could automate the tracing 
> and geneation of plots like this.  Some time I'd like to be able to 
> include tracing like this during CBT benchmark runs.
>
Thanks Mark! I'll try to play with CBT and tracing automation, and I'll 
let you know if I find anything interesting.

Mohamad