here's the wiki describing the system setup:
https://cwiki.apache.org/confluence/display/SPARK/Spark+QA+Infrastructure

we have 1 master and 8 worker nodes, 12 executors per worker (we'd be
better off w/more and smaller worker nodes however).

you don't need to install sbt -- it's in the build/ directory.

the pull request builder builds in parallel, but the master builds require
specific ports to be reserved and each build effectively locks down a
worker until it's done.  since we have 8 worker nodes, it's not *that* big
of a deal...

shane

On Tue, Feb 3, 2015 at 4:36 AM, scwf <wangf...@huawei.com> wrote:

> Here my question is:
>     1 How to set jenkins to make it build for multi PR parallel?. or one
> machine only support one PR building?
>     2 do we need install sbt on the CI machine since the script
> dev/run-tests will auto fetch the sbt jar ?
>
> - Fei
>
>
>
> On 2015/2/3 15:53, scwf wrote:
>
>> Hi, all
>>    we want to set up a CI env for spark in our team, is there any
>> reference of how to install jenkins over spark?
>>    Thanks
>>
>> Fei
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscr...@spark.apache.org
>> For additional commands, e-mail: dev-h...@spark.apache.org
>>
>>
>>
>>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscr...@spark.apache.org
> For additional commands, e-mail: dev-h...@spark.apache.org
>
>

Reply via email to