As said before, profiling is probably the way to go here. Good tools have 
been given. I personally prefer node --prof and inspect the results with 
v8profiler or https://www.npmjs.org/package/node-tick-processor
If you need finer grain on when you want to profile (for example, a single 
request) you can use https://www.npmjs.org/package/profiler

My guess, given an ecosystem of HTTPS and SOAP is that a lot of time is 
spent establishing secure sessions and parsing/serializing XML documents.

Of course, if your architecture allows it, move away from SOAP ;) you'd get 
caching for free in the protocol.

On Thursday, 29 May 2014 06:15:44 UTC+2, Reza Razavipour wrote:
>
> I have a RESTful node server. Clients send requests which all translate to 
> SOAP calls. The SOAP server and node server all both on the same machines.
> One HTTPS request, all are HTTPS, can translate into up to 7 SOAP calls. 
>
> When the client puts the auto refresh on, every 5 seconds a series of 
> requests are sent to the node server. With 2 clients, auto refresh on, can 
> add up to two dozens of SOAP calls,
> the node server, using top, goes up above 80%.
>
> How can I determine what is causing this CPU spike? 2 dozens of SOAP calls 
> can cause this? 
>
>
>

-- 
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/382cc919-2f89-4335-8293-36e02f90fb0f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to