Re: Hosting a Large Number of Node Apps on the IBM i

Hi Henrik,

Agreed, the Javascript runtime is single threaded. It is libuv that makes
threads happen. Here is an excerpt from the URL I forwarded earlier**.

"How the I/O is run in the background is not of our concern, but due to the
way our computer hardware works, with the thread as the basic unit of the
processor, libuv and OSes will usually run background/worker threads and/or
polling to perform tasks in a non-blocking manner."

Again, Javascript runtime is still single threaded BUT when it makes a call
to libuv, that libuv call could be to spawn a new thread within that same
process to do things that would cause the Javascript runtime to block. I
see Nathan has responded and he explains it better than I have.

**https://nikhilm.github.io/uvbook/basics.html


Aaron Bartell
litmis.com - Services for open source on IBM i


On Wed, Oct 14, 2015 at 11:49 AM, Henrik Rützou <hr@xxxxxxxxxxxx> wrote:

Aaron

what you are stating doesn't make sense. Javascript is known as a single
threaded
event driven and asynchronously (not to be confused with muli-threaded) OO
language
(Class or Prototype based).

Google's V8 javascript machine that runs in your chrome browser is exactly
the same that
the one that runs in node.js.

Putting a layer benieth it om a server that just executes instructions here
and there would
completely ruin the language and the JVM since you can't be sure that one
instruction in line
100 is processed before line 101 that may rely on result in line 100 is
finished.

On Wed, Oct 14, 2015 at 6:38 PM, Nathan Andelin <nandelin@xxxxxxxxx>
wrote:

I'm always open to other points of view.

There seems to be general agreement about external reverse-proxies /
load-balancers being an effective means of "routing" work to multiple
Node.js instances (Jobs), which ensures scalability and fail-over

support.

There doesn't seem to be much distinction whether one uses IIS, NGiNX,
Apache, or whatever for external request routing.

It would be nice to see some benchmarks pertaining to Node's internal
load-balancing via the "Cluster" module. I'd be a bit wary of it after

just

reading the documentation.

The question that has me the most stumped is the one about "routing"
potentially thousands of URLs to appropriate JavaScript routines, with

each

routine pertaining to various "applications" within "modules" within
"major-system-areas" (however you organize you broadly-scoped services)?

Kevin indicated that Node has to be restarted each time a routing
configuration changes in Sails. Inline-coding of routes in Express
obviously suffers from the same problem. I'd be concerned about

performance

as the number of URLs increase.
--
This is the Web Enabling the IBM i (AS/400 and iSeries) (WEB400) mailing
list
To post a message email: WEB400@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/web400
or email: WEB400-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/web400.

--
Regards,
Henrik Rützou

http://powerEXT.com <http://powerext.com/>
--
This is the Web Enabling the IBM i (AS/400 and iSeries) (WEB400) mailing
list
To post a message email: WEB400@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/web400
or email: WEB400-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/web400.