I for one am not a fan of x86 and run a range of GNU/Linux systems on PPC and other hardware platforms.

My brother Ed Blanchfield has infact blog’ed his efforts in running up Gentoo/Linux on some big iron I’ve loaned him, in the form of a Dual Proc Sun Microsystems SunFire E420R and a truck load of RAID stoage.

You can check out his efforts ( still to be completely detailed though ) over at:

http://www.e-things.org/

there’s a link to his Blog there. (check out his PI to the sound of MIDI as well while you’re there - it’s quite something !!)

What you’ll find if you do have the opportunity to run up the likes of either a GNU/Linux distro, or even one of the BSD’s, on heavy iron like Sun’s SPARC hardware, is that unlike the PC legacy performance issues of x86 platforms, you will be amazed by the performance difference of something like an E420R when you throw some serious load at the thing, throughput is the key here, forget about piddly little 500 of 800 Mhz FSB’s, try gigabit backplanes - crossbar at that, now you’re talking my language.

Databases for example, MySql or PostgreSQL, simply “fly” on hardware that’s designed from ground up to “pump data”, rather than hardware that’s been more and more tightly wound over the years to the point where under load you feel like it’s about to fly apart.

I’d put my money on even an old chunk of SPARC hardware over current x86 platforms

Sorry that this was way off thread but really had to get that off my chest

Cheers,

Dez
—
Dez Blanchfield
http://CradleTechnologies.COM
http://WebSearch.COM.AU
http://Mosman.no-ip.com/blog/

I’ve never used AIX, so I can’t comment on it directly, but I know I’ve personally contributed capability to a Linux project that has never been in AIX (IPC facilities for STREAMS). I don’t see it mentioned much how much more the open source community can achieve than can a single company, even IBM or Microsoft. IBM, I think, knows that it can’t compete in software, and likely doesn’t want to do so (it never seemed all that interested in software innovation even in the old days). It’s not even hardware that it really wants to sell; it wants to sell services. That’s also been true since the early days, and it true for other companies (like telecom providers) as well. Don’t cellphone companies practically give away phones (I don’t have one personally) to get customers to pay them on a monthly basis?

It works to the advantage of companies like IBM and Sun to put free software on their hardware and sell support for the combination. Among other things, it puts pressure on those companies (like Microsoft) whose sole business is selling software, and in the long term, software sales must be a losing proposition.

The point is, as I’m sure you’re aware, the question of what companies have to do to stay in business is not a simple one.

That is reflected in how companies hire. I think that now, because so much open source software is available for free, there aren’t going to be many job openings for original developers of proprietary software products. Thus, most posts will be for service-oriented positions instead, and developers are simply overqualified for most such positions.

That’s why it would be interesting, in my view, to try to break down by developer vs. non-developer. I realize it may be hard to do. But I would finally note that use of a programming language or facility with an OS don’t make this distinction. Scripting languages especially apply to both, but I’ve been around long enough to know that even hardcore C and assembly programming can be maintenance work predominantly.