Re: Why does node's Toolkit for i, use a database connection to execute CL commands?

It may or may not be of interest or relevant, but we have lots of code that sits and waits of data on a keyed data queue. It waits for 5 seconds then loops round and checks some stuff (like an instruction to end) then waits again.

When I did the same in a node app using itoolkit, the LPAR simply died on its feet. It wasn't obvious why because the node app was not showing up as using much CPU. However it certainly was the culprit - I changed to wait for 60 seconds instead as it wasn't essential to loop every 5 secs. The problem went away.

On 10 Oct 2015, at 15:58, Nathan Andelin <nandelin@xxxxxxxxx> wrote:

I am curious whether or not node scales more efficiently than the .NET
Provider in terms of concurrent users... I'm just curious which handles
concurrent users more efficiently in terms of CPU%.

All mainstream web application architectures scale by adding more cores and
memory and distributing (load balancing) "requests" across pools of
application server instances, and occasionally across pools of virtual
machine instances.

You're smart to ask about CPU efficiency, IMHO. And the other concern is
how much additional time system administrators will need to allocate to
"managing" distributed computing architectures.

Regarding CPU efficiency, you should have a wake-up call if you add 5250
replacement to the comparison. Any of the web application interfaces we've
mentioned in this thread (Java, PHP, Python, Ruby, Node.js, MS .Net) will
increase CPU usage by at least 3,000% over comparable 5250 interfaces.

5250 interfaces that consume 3 milliseconds of CPU time will consume a
minimum of 90 milliseconds after being converted to use a browser user
interface. Web interfaces often consume 200-300 milliseconds of CPU time to
generate "an HTML page".

The question of comparing CPU efficiency of Node.js vs. MS .Net is hard due
of the lack of benchmarks. Anecdotal evidence suggests that interpretive
scripting environments will consume several times more CPU than compiled
applications. But that could be ameliorated by reducing the amount of
"work" performed by Node.js or MS .Net by moving it to run in the IBM i
native environment.

Mike Pavlak of Zend has a story about a customer which was using XMLSERVICE
to run the ADDLIBLE command from PHP and complained about the performance.
Rather than "evoke" the "toolkit" for each library in the library list,
Mike suggested calling a CL program once to "add" all the needed libraries.
That fixed the performance problem.
--
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.


___________________________________________
This email has been scanned by iomartcloud.
http://www.iomartcloud.com/

________________________________

NOTICE: The information in this electronic mail transmission is intended by CoralTree Systems Ltd for the use of the named individuals or entity to which it is directed and may contain information that is privileged or otherwise confidential. If you have received this electronic mail transmission in error, please delete it from your system without copying or forwarding it, and notify the sender of the error by reply email or by telephone, so that the sender's address records can be corrected.



--------------------------------------------------------------------------------


CoralTree Systems Limited
25 Barnes Wallis Road
Segensworth East, Fareham
PO15 5TT

Company Registration Number 5021022.
Registered Office:
12-14 Carlton Place
Southampton, UK
SO15 2EA
VAT Registration Number 834 1020 74.