[Phpwm] hyperphp

[Tim Langley]

Whilst not "HyperPHP" there are several "PHP Compilers"
in particular Quercus (runs PHP on the JVM) is particularly impressive

http://www.caucho.com/resin-3.0/quercus/

T

On 12 Jan 2010, at 14:40, Richard Cunningham wrote:

> I don't think they are trying to scale PHP as such, they are trying to use less hardware so the costs are lower. Facebook's goals seem to be to reduce costs in general, like they did with the memcache binary protocol [reducing CPU load] and Cassandra [replacing netapp], because they don't bring that much in really (about $2 per user per year I think).
> 
> Also if the interview is real and I think it is, it's a bit old because the quoted Facebook user numbers are only 220 million and they are 350 million now. Therefore HyperPHP is either late or they haven't announced it (I assume they will open source it).
> 
> David Caunt wrote:
>> 
>> I'd expect a compiled application to be faster than an opcode cache (which is intermediate) but I have doubts about the existence of hyperphp. That said, Facebook have contributed to memcache and other open source projects in the past. Given that most of Facebook is cached, it seems unlikely that they would try to optimise a language. Languages don't scale!
>> 
>> Dave
>> 
>> 2010/1/12 Stephen Orr <steve at stephenorr.co.uk>
>> I would agree with you there, and it would produce similar performance gains. I suspect the interview was either faked, or that the employee was misadvised as to what the project was about - and most companies come up with clever names for internal projects... David will know exactly what I mean if I say the phrase "NewNet"...
>> 
>> Thanks
>> Steve
>> 
>> 2010/1/12 David Goodwin <david at codepoets.co.uk>
>> 
>> j michaelson wrote:
>> > http://therumpus.net/2010/01/conversations-about-the-internet-5-anonymous-facebook-employee/?full=yes
>> >
>> > An interview with a facebook employee. The guy mentions that someone
>> > there is making hyperphp, which is a compiled language instead of
>> > scripted. (mentioned about half way down the page).
>> >
>> 
>> I don't know whether that interview is fake or real... doesn't APC (and
>> other op-code caches) make the point of 'hyperphp' (compiled 'php')
>> redundant?
>> 
>> Perhaps I'm missing something.
>> 
>> David.
>> --
>> David Goodwin
>> 
>> [ david at codepoets dot co dot uk ]
>> [ http://www.codepoets.co.uk       ]
>> 
>> _______________________________________________
>> Phpwm mailing list
>> Website : http://www.phpwm.org
>> Twitter : http://www.twitter.com/phpwm
>> Facebook: http://www.facebook.com/group.php?gid=2361609907
>> 
>> Post to list: Phpwm at mailman.lug.org.uk
>> Archive etc : https://mailman.lug.org.uk/mailman/listinfo/phpwm
>> 
>> 
>> _______________________________________________
>> Phpwm mailing list
>> Website : http://www.phpwm.org
>> Twitter : http://www.twitter.com/phpwm
>> Facebook: http://www.facebook.com/group.php?gid=2361609907
>> 
>> Post to list: Phpwm at mailman.lug.org.uk
>> Archive etc : https://mailman.lug.org.uk/mailman/listinfo/phpwm
>> 
>> 
>> 
>> -- 
>> Save Water...Drink Ouzo !
>> 
>> _______________________________________________
>> Phpwm mailing list
>> Website : http://www.phpwm.org
>> Twitter : http://www.twitter.com/phpwm
>> Facebook: http://www.facebook.com/group.php?gid=2361609907
>> 
>> Post to list: Phpwm at mailman.lug.org.uk
>> Archive etc : https://mailman.lug.org.uk/mailman/listinfo/phpwm
> _______________________________________________
> Phpwm mailing list
> Website : http://www.phpwm.org
> Twitter : http://www.twitter.com/phpwm
> Facebook: http://www.facebook.com/group.php?gid=2361609907
> 
> Post to list: Phpwm at mailman.lug.org.uk
> Archive etc : https://mailman.lug.org.uk/mailman/listinfo/phpwm

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.lug.org.uk/pipermail/phpwm/attachments/20100112/78fffc34/attachment.htm