Hi All,

We ran a fail-over test case for queues with the implementation. Following
are the results :

*With RDBMS Slot Implementation : *
Consumer TPS : 1207.8700542930524
Average Latency : 2.7297796940344456
Consumer messages : 297223
Publisher TPS : 1224.0601258333809
Publisher messages : 300000

*With Hazelcast *:
Consumer TPS : 1201.9930300158992
Average Latency : 6.485580620486475
Consumer messages : 299378
Publisher TPS : 1210.0238374695982
Publisher messages : 300000

Notes :
The message loss can be considered due to the killing of the disruptor
buffer in andes.

Specs :
 - Message Store : MySQL.
 - Ran on a MacBook Pro.

Regards,
Hemika


Hemika Kodikara
Software Engineer
WSO2 Inc.
lean . enterprise . middleware
http://wso2.com

Mobile : +94777688882

On Mon, Apr 20, 2015 at 10:43 AM, Anuja Herath <[email protected]> wrote:

> Hi All,
>
> I have tested RDBMS slot implementation with test cases already there in
> "tests-clustering". Some test cases failed for both Hazelcast and RDBMS
> implementations and other test cases executed successfully for both. I have
> noted results in [1].
>
> What can be other scenarios that i need to implement, which will help
> to validate functionality of RDBMS Slot implementation?
>
> [1]
> https://docs.google.com/a/wso2.com/spreadsheets/d/1ejl6Ng5adoDYAO-gzLQ3NlEAINMKW92CixPeYJVoQz8/edit?usp=sharing
>
> Thanks,
> Anuja.
>
> --
> Anuja Herath
> *Software Engineer*
> *WSO2, Inc.*
> Mobile : +94 (0)71 429 8861
>
_______________________________________________
Dev mailing list
[email protected]
http://wso2.org/cgi-bin/mailman/listinfo/dev

Reply via email to