+ using strong cluster control, you can add dynamically additional
processes to a running application based on workload and drop them when not
needed.

You can avoid application restarts.

So if you started with 2 processes as part of the cluster and found them
overwhelmed with load, you can automatically resize the cluster to say 6
processes and the master will distribute load to the new processes.

Hope this helps.

Shubhra


On Thu, Dec 19, 2013 at 11:57 AM, Shubhra Kar <[email protected]> wrote:

> The awareness between processes is useful for work load management between
> master and child. Description by Alex is mostly accurate.
>
> We are also working on a more distributed work load management solution in
> strong-cluster to allocate specific type of workloads selectively to child
> processes. So you can manage it like a job and individual tasks can be
> split and can run dedicated on a specific set of processes in the cluster.
>
> Cluster awareness is useful for sharing state and session information
> across processes. Not every node application will need it, definitely not
> stateless ones. So could look like over-engineering. But in cases where
> transaction integrity and rollback are needed - say banking transactions
> (if one processes drops a transaction in flight due to load, maybe the
> other can pick it up), it becomes critical.
>
> Kind Regards,
> Shubhra
>
>
>
> On Thursday, December 19, 2013 10:41:22 AM UTC-8, Bijuv V wrote:
>
>> Thanks for sharing your views . Do you know what Ryan means by Server
>> File Descriptors?
>>
>> Also , about Strongloop, you mentioned about communication between the
>> node instances - why would this be necessary - cant figure out a use case .
>>
>>
>> On Thu, Dec 19, 2013 at 6:28 PM, Luke Arduini <[email protected]>wrote:
>>
>>> The "experimental" stability index means almost nothing. It's fine.
>>> Plenty of people use it in production.
>>>
>>>
>>> On Thursday, December 19, 2013, Matt wrote:
>>>
>>>>
>>>> On Thu, Dec 19, 2013 at 11:50 AM, Bijuv V <[email protected]> wrote:
>>>>
>>>>> a. Use Cluster feature of node. Everything is handled by node. Still
>>>>> experimental AFAIK.
>>>>>
>>>>
>>>> I think while cluster is still marked experimental, it's built on top
>>>> of child_process which is marked as Stable.
>>>>
>>>> Most people go with cluster. It works well and is extremely reliable in
>>>> my experience. As long as you're aware of the gotchas with graceful
>>>> restarts.
>>>>
>>>> Matt.
>>>>
>>>> --
>>>> --
>>>> Job Board: http://jobs.nodejs.org/
>>>> Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-
>>>> Posting-Guidelines
>>>> You received this message because you are subscribed to the Google
>>>> Groups "nodejs" group.
>>>> To post to this group, send email to [email protected]
>>>> To unsubscribe from this group, send email to
>>>> [email protected]
>>>> For more options, visit this group at
>>>> http://groups.google.com/group/nodejs?hl=en?hl=en
>>>>
>>>> ---
>>>> You received this message because you are subscribed to the Google
>>>> Groups "nodejs" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>> an email to [email protected].
>>>>
>>>> For more options, visit https://groups.google.com/groups/opt_out.
>>>>
>>>  --
>>> --
>>> Job Board: http://jobs.nodejs.org/
>>> Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-
>>> Posting-Guidelines
>>> You received this message because you are subscribed to the Google
>>> Groups "nodejs" group.
>>> To post to this group, send email to [email protected]
>>>
>>> To unsubscribe from this group, send email to
>>> [email protected]
>>>
>>> For more options, visit this group at
>>> http://groups.google.com/group/nodejs?hl=en?hl=en
>>>
>>> ---
>>> You received this message because you are subscribed to a topic in the
>>> Google Groups "nodejs" group.
>>> To unsubscribe from this topic, visit https://groups.google.com/d/
>>> topic/nodejs/Dv5J7w34rcQ/unsubscribe.
>>> To unsubscribe from this group and all its topics, send an email to
>>> [email protected].
>>>
>>> For more options, visit https://groups.google.com/groups/opt_out.
>>>
>>
>>  --
> --
> Job Board: http://jobs.nodejs.org/
> Posting guidelines:
> https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
> You received this message because you are subscribed to the Google
> Groups "nodejs" group.
> To post to this group, send email to [email protected]
> To unsubscribe from this group, send email to
> [email protected]
> For more options, visit this group at
> http://groups.google.com/group/nodejs?hl=en?hl=en
>
> ---
> You received this message because you are subscribed to a topic in the
> Google Groups "nodejs" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/nodejs/Dv5J7w34rcQ/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> [email protected].
> For more options, visit https://groups.google.com/groups/opt_out.
>



-- 
*Shubhra Kar | Director - Products & Systems Engineering  |  StrongLoop*
*Direct: 1-650-740-4723 | *[email protected]

-- 
-- 
Job Board: http://jobs.nodejs.org/
Posting guidelines: 
https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
You received this message because you are subscribed to the Google
Groups "nodejs" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/nodejs?hl=en?hl=en

--- 
You received this message because you are subscribed to the Google Groups 
"nodejs" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to