[lttng-dev] Performance impact using the "filter" option

[Amit Margalit]

If I understand correctly, filtering is done at the client application 
side.

This means that filtering could theoretically lower the performance of the 
running application which has to check, for each event, whether it should 
be sent to the daemon or not.

Naturally, if you filter some events, the chances of losing events due to 
congestion at the daemon are lower.

Amit Margalit
IBM XIV - Storage Reinvented
XIV-NAS Development Team
Tel. 03-689-7774
Fax. 03-689-7230



From:   Ilya Mirsky <ilya.mirsky at gmail.com>
To:     lttng-dev at lists.lttng.org
Date:   03/19/2014 02:47 PM
Subject:        [lttng-dev] Performance impact using the "filter" option



Hi all,

It's a simple question, but I couldn't find the answer in the 
documentation.
Figured out I'll ask here before looking this in the code.

Should I expect LTTng's overhead to be lower when filtering some of the 
trace points,
or is the main overhead is due to the trace point call and not the 
recording itself?
In practice I see no noticeable difference in performance, 
while I was hoping to get at least some improvement...

Thanks,
Ilya



_______________________________________________
lttng-dev mailing list
lttng-dev at lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lttng.org/pipermail/lttng-dev/attachments/20140319/18df8edb/attachment.html>