[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: [gyrex-dev] Scaling of Gyrex Jobs
|
Hi Jürgen.
Given the current implementation this will depend heavily on your ZooKeeper cluster. The current implementation would put a lot write activity on your ZooKeeper cluster because the job history is stored in ZooKeeper. In the past we've discussed implementing an alternate ZooKeeper history storage location. However, that hasn't been implemented yet.
Depending on what your requirements are, you may also benefit from a volatile job, i.e. one that doesn't store history at all.
-Gunnar
--
Gunnar Wagenknecht
gunnar@xxxxxxxxxxxxxxx, http://guw.io/
> Am 06.12.2015 um 23:42 schrieb Jürgen Albert <j.albert@xxxxxxxxxxxxxxxxxx>:
>
> Hi,
>
> does anyone have any idea how good the gyrex jobs can handle large amounts of tiny jobs?
>
> We have a case with an estimate of 10000+ jobs, which are in itself not that much time consuming. Point is, we have to be able to repeat them like every 3 minutes (not all at the same time) in a cluster that will be sized appropriately. The question is, if there is some experience, if the implementation can handle such a load. In the future we might be facing over 30000 Jobs.
>
> Thanks heaps,
>
> Jürgen.
>
> --
> Jürgen Albert
> Geschäftsführer
>
> Data In Motion UG (haftungsbeschränkt)
>
> Kahlaische Str. 4
> 07745 Jena
>
> Mobil: 0157-72521634
> E-Mail: j.albert@xxxxxxxxxxxxxxx
> Web: www.datainmotion.de
>
> XING: https://www.xing.com/profile/Juergen_Albert5
>
> Rechtliches
>
> Jena HBR 507027
> USt-IdNr: DE274553639
> St.Nr.: 162/107/04586
>
> _______________________________________________
> gyrex-dev mailing list
> gyrex-dev@xxxxxxxxxxx
> To change your delivery options, retrieve your password, or unsubscribe from this list, visit
> https://dev.eclipse.org/mailman/listinfo/gyrex-dev
