You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@hadoop.apache.org by Rohan Rai <ro...@inmobi.com> on 2010/05/11 15:57:55 UTC

JVM Reuse for parallel execution

Hi

Is there a way of executing multiple tasks in parallel in a single JVM.

For a particular job we see a child process being spawned for each task,
despite
mapred.job.reuse.jvm.num.tasks being set to -1

In order to avoid the overhead of starting a JVM for each task and yet
achieve
parallel execution of tasks, wanted to know sharing the JVM is possible, for
parallel task execution.

Regards
Rohan

The information contained in this communication is intended solely for the use of the individual or entity to whom it is addressed and others authorized to receive it. It may contain confidential or legally privileged information. If you are not the intended recipient you are hereby notified that any disclosure, copying, distribution or taking any action in reliance on the contents of this information is strictly prohibited and may be unlawful. If you have received this communication in error, please notify us immediately by responding to this email and then delete it from your system. The firm is neither liable for the proper and complete transmission of the information contained in this communication nor for any delay in its receipt.

Re: JVM Reuse for parallel execution

Posted by Tim Robertson <ti...@gmail.com>.
http://hadoop.apache.org/common/docs/current/mapred_tutorial.html#Task+JVM+Reuse?

(Just so you know - there is a mapreduce users mailing list for MR specific
questions)

Hope this helps,
Tim

On Tue, May 11, 2010 at 4:02 PM, Rohan Rai <ro...@inmobi.com> wrote:

> Forgot to add, we are using the following configuration.
>
> mapred.tasktracker.map.tasks.maximum as 4
> and mapred.tasktracker.reduce.tasks.maximum as 3.
>
> Regards
> Rohan
>
>
>
> Rohan Rai wrote:
>
>> Hi
>>
>> Is there a way of executing multiple tasks in parallel in a single JVM.
>>
>> For a particular job we see a child process being spawned for each
>> task, despite
>> mapred.job.reuse.jvm.num.tasks being set to -1
>>
>> In order to avoid the overhead of starting a JVM for each task and yet
>> achieve
>> parallel execution of tasks, wanted to know sharing the JVM is
>> possible, for
>> parallel task execution.
>>
>> Regards
>> Rohan
>>
>>
>
> The information contained in this communication is intended solely for the
> use of the individual or entity to whom it is addressed and others
> authorized to receive it. It may contain confidential or legally privileged
> information. If you are not the intended recipient you are hereby notified
> that any disclosure, copying, distribution or taking any action in reliance
> on the contents of this information is strictly prohibited and may be
> unlawful. If you have received this communication in error, please notify us
> immediately by responding to this email and then delete it from your system.
> The firm is neither liable for the proper and complete transmission of the
> information contained in this communication nor for any delay in its
> receipt.
>

Re: JVM Reuse for parallel execution

Posted by Rohan Rai <ro...@inmobi.com>.
Forgot to add, we are using the following configuration.

mapred.tasktracker.map.tasks.maximum as 4
and mapred.tasktracker.reduce.tasks.maximum as 3.

Regards
Rohan


Rohan Rai wrote:
> Hi
>
> Is there a way of executing multiple tasks in parallel in a single JVM.
>
> For a particular job we see a child process being spawned for each
> task, despite
> mapred.job.reuse.jvm.num.tasks being set to -1
>
> In order to avoid the overhead of starting a JVM for each task and yet
> achieve
> parallel execution of tasks, wanted to know sharing the JVM is
> possible, for
> parallel task execution.
>
> Regards
> Rohan
>


The information contained in this communication is intended solely for the use of the individual or entity to whom it is addressed and others authorized to receive it. It may contain confidential or legally privileged information. If you are not the intended recipient you are hereby notified that any disclosure, copying, distribution or taking any action in reliance on the contents of this information is strictly prohibited and may be unlawful. If you have received this communication in error, please notify us immediately by responding to this email and then delete it from your system. The firm is neither liable for the proper and complete transmission of the information contained in this communication nor for any delay in its receipt.