storm-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Nathan Marz <nat...@nathanmarz.com>
Subject Re: How to evaluating latency per tuple?
Date Tue, 04 Feb 2014 08:52:44 GMT
1 is already calculated by Storm and reported in default metrics. For 2 you
could track that yourself in your spout implementation.


On Mon, Feb 3, 2014 at 7:40 PM, Sho Nakatani <lay.sakura@gmail.com> wrote:

> Hi all,
>
> I'm new to Storm.
> I'm finding the way to count latency per tuple.
>
> What I exactly want are:
> (1) Mean latency of all tuples elapsed between a Spout generates them
> and last Bolt processes them
> (2) Maximum latency of all tuples elapsed between a Spout generates
> them and last Bolt processes them
>
> Example:
> If topology is
>   Spout => Bolt0 => Bolt1
> and tupleA is generated on 12:00:00 at Spout and finished on 12:00:05
> at Bolt1, the latency of tupleA is 5 seconds.
>
> Especially, I want to calculate latencies of WourdCountTopology in
> storm-starter.
>
> Any advice?
> Thanks,
>
> Nakatani
>



-- 
Twitter: @nathanmarz
http://nathanmarz.com

Mime
View raw message