Gail Shaw wrote:

>> A surprising amount of software
>> still uses such brain-dead technology. (E.g., I frequently see it in
>> printer and scanner drivers. Like my mum's HP printer, that connects
>> over the parallel port. Printing to it causes 100% CPU usage until the
>> print finishes. No reason for it...)
> 
> I've seen exactly the same thing on a piece of hardware I had to deal with.
> Can't remember the name of it (Rakal or something similar)
> Drove one CPU's kernal usage to 100% flat. Lovely thing to have on a
> database server.

Yes, truly wonderful, isn't it?

OOC, how many cores does your server have? (Mine has 2. And they're both 
Pentium-III...)

>> I just wonder if one day it will be, like, you run a program and as many
>> computers as you have switched on automatically start biting into it...
>> but hey, that sounds *far* too easy! ;-) Even Apple haven't done it yet.
> 
> I don't think it's that farfetched. You could argue that Google's done
> similar with their distributed search.

I'm not familiar with that.

> Distributed computing was a big research topic when I was at iuniversity. I
> think it still is. I keep running across references when I'm reading
> It's not easy though.

The Haskell guys keep telling us how parallel processing is the future 
and how Haskell is right there ready to step in... but it's not actually 
that easy. Haskell has far fewer problems than *cough* C. Emphasis fewER 
problems...

Anyway, things like PVM exist. But it's nowhere near as easy as opening 
Word and start typing... You've got to configure a whole heap of stuff 
to get it going, and then you can't just *run* your program, you must 
configure that also. And your program must be specially built to work 
with PVM, and... and...

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message