You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@geode.apache.org by Bruce Schuchardt <bs...@pivotal.io> on 2018/08/22 16:39:38 UTC
tests are using too much memory
The single-threaded gradle build is now using xmx3g but when tests start
running it's consuming over 4gb. I also have Intellij and Chrome
running on my 16gb laptop and the tests are causing me to run out of
RAM. It's set to use 1/2 of the CPUs on a machine but I am finding that
all of my CPUS are pegged.
Can we change this behavior to only run tests in parallel if "-parallel"
has been specified, or ramp it down to 1/4 of the CPUs?
Re: tests are using too much memory
Posted by Jacob Barrett <jb...@pivotal.io>.
https://github.com/apache/geode/pull/2362
On Wed, Aug 22, 2018 at 10:21 AM Jacob Barrett <jb...@pivotal.io> wrote:
> https://issues.apache.org/jira/browse/GEODE-5620
>
>
> On Wed, Aug 22, 2018 at 9:39 AM Bruce Schuchardt <bs...@pivotal.io>
> wrote:
>
>> The single-threaded gradle build is now using xmx3g but when tests start
>> running it's consuming over 4gb. I also have Intellij and Chrome
>> running on my 16gb laptop and the tests are causing me to run out of
>> RAM. It's set to use 1/2 of the CPUs on a machine but I am finding that
>> all of my CPUS are pegged.
>>
>> Can we change this behavior to only run tests in parallel if "-parallel"
>> has been specified, or ramp it down to 1/4 of the CPUs?
>>
>>
Re: tests are using too much memory
Posted by Jacob Barrett <jb...@pivotal.io>.
https://issues.apache.org/jira/browse/GEODE-5620
On Wed, Aug 22, 2018 at 9:39 AM Bruce Schuchardt <bs...@pivotal.io>
wrote:
> The single-threaded gradle build is now using xmx3g but when tests start
> running it's consuming over 4gb. I also have Intellij and Chrome
> running on my 16gb laptop and the tests are causing me to run out of
> RAM. It's set to use 1/2 of the CPUs on a machine but I am finding that
> all of my CPUS are pegged.
>
> Can we change this behavior to only run tests in parallel if "-parallel"
> has been specified, or ramp it down to 1/4 of the CPUs?
>
>