Thanks Thomas!!

Paul, I looked at all 3 and really liked what Thomas sent (
https://dropwizard.github.io/metrics/3.1.0/getting-started/). Licence is
Apache 2.

I think I will try this out with a training project.

--Srinath

On Sat, Sep 27, 2014 at 11:37 AM, Thomas Wieger <[email protected]>
wrote:

> please have a look on the metrics lib, which quite nicely provides support
> for capturing different kinds of statistics. see
> https://dropwizard.github.io/metrics/3.1.0/
>
>
> regards,
>
> thomas
>
> Am Freitag, 26. September 2014 schrieb Sajith Ravindra :
>
>
>> IMHO, CEP is another product where this kind of probes can  be really
>> useful as we have to deal with performance related requirements. As there's
>> a major refactoring going on in siddhi I think we can we plant probes like
>> this into the Siddhi engine. It will be useful to evaluate the re-factored
>> Sindhi engines performance as well as to figure out performance bottlenecks
>> in the future.
>>
>>
>> Thanks
>> *,Sajith Ravindra*
>> Senior Software Engineer
>> WSO2 Inc.; http://wso2.com
>> lean.enterprise.middleware
>>
>> mobile: +94 77 2273550
>> blog: http://sajithr.blogspot.com/
>> <http://lk.linkedin.com/pub/shani-ranasinghe/34/111/ab>
>>
>> On Fri, Sep 26, 2014 at 8:38 AM, Srinath Perera <[email protected]> wrote:
>>
>>> We have some pretty nice probes already in ESB, that let you look into
>>> what is going in the same running in production.  I think we need to do
>>> this generally and build it to other products. That should reduce time we
>>> spent debugging issues significantly.
>>>
>>> My proposal is to build a probe Lib that looks like following.
>>>
>>> There are two kinds of things you need to collect.
>>>
>>>    1. Throughput at give point of code (how fast calls goes though)
>>>    2. Latency between two points.
>>>
>>>
>>> We will have two types of Probes. Code would look like following.
>>>
>>> Probe probe = new Probe("Name", "throughput", timeDuration);
>>> ..
>>> probe.recordThroughput();
>>>
>>> Probe probe = new Probe("Name", "latency", timeDuration, logLevel);
>>> ...
>>> long id = probe.startTicking() // this so same probe can be used by many
>>> threads
>>> ...
>>> probe.endTicking(id);
>>>
>>> Probe will summarise data over given duration and expose. We need
>>>
>>>
>>>    1. JMX bean
>>>    2. BAM Agent
>>>    3. Can turn on, off via JMX agent or via System property
>>>    4. Have Tool Box
>>>    5. Have in product UI
>>>    6. Can configured to write data to logs
>>>
>>> Each probe should be very small and should be able to create thousands
>>> without much effect. (e.g. create one for each mediator type)
>>>
>>> WDYT?
>>>
>>> --Srinath
>>>
>>>
>>>
>>>
>>>
>>>
>>> --
>>> ============================
>>> Blog: http://srinathsview.blogspot.com twitter:@srinath_perera
>>> Site: http://people.apache.org/~hemapani/
>>> Photos: http://www.flickr.com/photos/hemapani/
>>> Phone: 0772360902
>>>
>>> _______________________________________________
>>> Architecture mailing list
>>> [email protected]
>>> https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture
>>>
>>>
>>


-- 
============================
Blog: http://srinathsview.blogspot.com twitter:@srinath_perera
Site: http://people.apache.org/~hemapani/
Photos: http://www.flickr.com/photos/hemapani/
Phone: 0772360902
_______________________________________________
Architecture mailing list
[email protected]
https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture

Reply via email to