You are viewing a plain text version of this content. The canonical link for it is here.

Posted to user@geode.apache.org by "Xu, Nan" <nx...@baml.com> on 2017/06/20 13:42:53 UTC

subscripting client slow down server publish?

Hi,

     I have a simple setup, 1 publisher (proxy client), 1 subscriiber(proxy client), 1 server. publisher publish at full speed and about 12,000 message/s, after client start to subscribe(no durable, no conflation ), I see the publish speed drop to about 7,000. Is this expected?

Thanks,
Nan

----------------------------------------------------------------------
This message, and any attachments, is for the intended recipient(s) only, may contain information that is privileged, confidential and/or proprietary and subject to important terms and conditions available at http://www.bankofamerica.com/emaildisclaimer.   If you are not the intended recipient, please delete this message.

Re: subscripting client slow down server publish?

Posted by Anthony Baker <ab...@pivotal.io>.

It’s hard to speak in absolute numbers.  Certainly I have seen sustained throughput performance easily exceeding 10X on clusters with multiples clients and servers.  Drops in throughput can be caused by GC, cpu contention, network contentions, ...

Anthony

> On Jun 20, 2017, at 6:42 AM, Xu, Nan <nx...@baml.com> wrote:
> 
> Hi,
>  
>      I have a simple setup, 1 publisher (proxy client), 1 subscriiber(proxy client), 1 server. publisher publish at full speed and about 12,000 message/s, after client start to subscribe(no durable, no conflation ), I see the publish speed drop to about 7,000. Is this expected?
>  
> Thanks,
> Nan
> This message, and any attachments, is for the intended recipient(s) only, may contain information that is privileged, confidential and/or proprietary and subject to important terms and conditions available at http://www.bankofamerica.com/emaildisclaimer <http://www.bankofamerica.com/emaildisclaimer>. If you are not the intended recipient, please delete this message.