NYCPHP Meetup

[inforequest]

Sol sol2ray-at-yahoo.fr |nyphp 04/2004| wrote:

>>Of course, I'm on a cheap shared host, but how hard
>>is it to serve static
>>content? :-)
>>
>>    
>>
>>From Chris' website:
>"I think PHP scales well because Apache scales well
>because the Web scales well. PHP doesn't try to
>reinvent the wheel; it simply tries to fit into the
>existing paradigm, and this is the beauty of it."
>-----------------
>Sorry for bringing this back but I still don't
>understand this "SCALABILITY" thing. The more I read
>you guyz, the more I get confused about the whole
>thing.
>
>Can someone give me a more "IDIOT PROOF" explanation? 
>  
>

       George Wrote:
              "The modularity of the system is completely orthogonal to 
the point."


Hah hah.. George I think he was asking for simpler... and besides it is 
not possible to be orthogonal to a point... you need a plane ;-)

Anyways by my read Chris was offloading responsibility for his blog not 
"sclaing well" or (not serving his pages well as the demand increased), 
because serving the pages is a job of his web host's servers. They 
obviously weren't configured to "scale well" with respect to Chris' 
blog, but that is not a function of Chris pages (which he tells are not 
even PHP - they are prepared static "images" of his pages).

So now that we understand Chris' site going down has no relation to his 
comments about the scalability of PHP. What about those comments?

PHP is a scripting engine -- it runs code. If the PHP engine is asked to 
run alot of code, does it perform about as well as it did when it was 
asked to run a little code? Herein lies the problem.. is the "quantity" 
of code a matter of server requests or complexity of script? Hard to 
say. Perhaps Chris refers to complexity of code, such that code 
fragments are re-used (functions, modules, objects, libraries, 
whatever). Does PHP lend itself to neat and efficient organization of 
those "fragments" such that as programs get complex, they stay organized 
well enough for the rest of the components (server, Apache, cache, etc) 
to continue to perform well? What about shared memory, as in variables? 
As they are shared, does performance continue at similar levels or does 
the system slow down with increased shared demand?

Hard to say if these aspects are of ultimate importance.... but 
certainly they are all part of the picture, no?

So perhaps the argument of "scalability of PHP" is best addressed 
through empirical examples.... if it is being successfully deployed in 
scaled applications by happy customers. or perhaps even "evangelically 
happy(TM)" customers, then can we say it is scalable.. or perhaps we can 
simply say scalability doesn't seem to be an issue?

It is obvious that the BASIC computer language did not scale well. It is 
obvious that the COBOL computer language did scale well. It is obvious 
that Windows does not scale well, when compared to Linux. It is obvious 
that MS Access does not scale well, especialy when compared to any other 
database on the face of the earth. It seems Chris was driven to comment 
on the fact that someone who needed an application that could scale well 
had abandoned one tchnology (based on empirical judgement) and adopted a 
PHP solution. perhaps highlioghting the adoption and success is the most 
important thing that can be done to make an argument for the 
"scalability" of PHP.


--=john
John F Andrews
(insert commercial signature lines here)

(Disclaimer: I do not own any M$ stock, nor do I recommend it to my 
clients).