[App_rpt-users] *** NODE LOAD TESTING HAS BEEN CANCELED FOR THIS SAT 14th ***

[Mike]

***   NODE LOAD TESTING HAS BEEN CANCELED FOR THIS SAT 14th   ***

Reason...

'One' participant, and that participant stated he would not be around 
for the test. ????????

While I think this is 'very important',

I personally will not 'ENTERTAIN'  anyone state to me that I should use 
a 'untested resource' for use in  nets etc .

Nobody will be putting your head on the chopping block when things do 
not work out.

I feel the more the better but _everyone needs some experience doing 
this sort of stuff. _Pass the torch !_
_

And this is not the same game it was just 2-3 years ago. I do know since 
I have taken part in many very large nets.

How do you make judgements on load distribution without this data ?    
Just throw a lot of what you think are large assets at it and 'hope' it 
works ?

_Well, Good luck in your 'RELIABLE' ecomm endeavors. _

Anyone think they have a 'magic formula' for figuring this out ? ...   
Keep guessing !

Most think it is simply a Network Bandwidth issue, and more often, it is 
a CPU/Memory bandwidth issue.

Allison will eat into that 'heavily on smaller systems'. Heavy loads to 
your allmon page will do the same because it uses PHP server side 
scripting and is constantly querying your system for live data with each 
visitor that just sits there and monitors. Unlike HTML where you are 
just 'reading' the text, PHP uses the servers CPU to generate a HTML 
output in a dynamic way. I could state other things that eat into your 
available mem/cpu but for most of us, those are the 2 major ones but 
does not account for other services some run in their systems.

Both of those mean less available resources for voice data to be 
distributed. So it is not just about 'Internet Bandwidth'.

(It's-A Put-Up or Shut-Up situation.  Just so all those 'back seat 
drivers' know where they stand with me.)

I have been putting the work up to make 'even better things' possible 
and understandable going forward, and most of all, PLAN-ABLE.


A large group of successful 'real world' tests will help us develop a 
formula that everyone will benefit from.

Or we can continue to head down the path of failure from our own 
ignorance as this continues to grow rapidly.

The issues will not be going away, but getting more important.

I have stated my case for it the best I know how.

Blaming it on 'the summer', *_I will re-visit the idea again this winter_*.

73,

...mike/kb8jnm

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.keekles.org/pipermail/app_rpt-users/attachments/20180712/4effab9e/attachment.html>