ignite-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Vladimir Ozerov <voze...@gridgain.com>
Subject Re: Cache Connection taking time while using IgniteRDD on Spark executor
Date Thu, 28 Apr 2016 08:50:02 GMT
Hi Vij,

For now - yes, I would say it is expected startup time because Ignite has
to perform some actions to join the cluster. On the other hand, I agree
with Valentin, that this definitely should be improved for client
connections. I would say that <1s would be a good number for me.

Vladimir.

On Thu, Apr 28, 2016 at 9:00 AM, vijayendra bhati <veejayendraa@yahoo.com>
wrote:

> But in the mean while can I do something to improve these timings - is 4-5
> sec is usual time for start up ?
>
> Regards,
> Vij
>
>
> On Thursday, April 28, 2016 6:45 AM, vkulichenko <
> valentin.kulichenko@gmail.com> wrote:
>
>
> Hi Vij,
>
> Ignite starts full-fledged node on the client side to provide the whole set
> of available APIs. This makes the startup process a bit heavy, but there is
> a room for optimizations. There is a ticket for this that you can track:
> https://issues.apache.org/jira/browse/IGNITE-3060
>
> -Val
>
>
>
> --
> View this message in context:
> http://apache-ignite-users.70518.x6.nabble.com/Cache-Connection-taking-time-while-using-IgniteRDD-on-Spark-executor-tp4601p4640.html
> Sent from the Apache Ignite Users mailing list archive at Nabble.com.
>
>
>
>

Mime
View raw message