Hi,
 

> We are trying to cluster a level deeper than is usual.
>
> NodeJS seems to developed to be able to run a code than can be accessed by 
> multiple users and every instance can be run by a different worker.
>

IMHO Node.JS can handle particular request (be it user, be it some URL 
route, etc.) by a different worker


> Now we have a case, where we have only one user with a very complicated 
> backend process. This single instance of the code causes the CPU to 
> overload at times so we want to split up the work over different CPU cores. 
>
 

> We have tried multiple approached but haven't been able to make it work 
> yet, any insight to this?
>
> What are they?

If that user doing many requests to particular URL of your app and those 
request one by one eat CPU very much, then maybe you can handle that URL by 
own cluster running on another TCP port and have nginx as router pointing 
to that URL? There must be some splitting boilerplate for that api 
controllers (js files) to load config, connect to db, etc as usual app 
does. 

Boby Rahmawan says "split into child process". Maybe it is the same.

-- 
Job board: http://jobs.nodejs.org/
New group rules: 
https://gist.github.com/othiym23/9886289#file-moderation-policy-md
Old group rules: 
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 unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/nodejs/bf574c56-959a-4031-a0f6-166b5645a49f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to