On Feb 6, 7:52 pm, Roy Schestowitz <newsgro...@xxxxxxxxxxxxxxx> wrote:
> Best Linux Distro for Mac
> ,----[ Quote ]
> | At the moment, I have a short list of Linux distros that could replace
> | Xubuntu Hardy on my Mac:
> | Fedora 10
> | I dual-booted Fedora 9 and Xubuntu before but ultimately decided to just use
> | Xubuntu since it performed a lot better than Fedora. Right now, I’m keeping
> | my fingers crossed that Fedora 10 will do wonders for my Mac.
> | Mepis 8.0
> | I have great memories using Mepis Linux, so I want to take the plunge and
> | test the upcoming version.
> | Sabayon Linux 4
> | Putting this Gentoo-based distro on my Mac can be suicide since I’m probably
> | on my own when I do this. But since danger is my middle name, I will most
> | likely try Sabayon Linux on my Macbook Pro.
The great thing about Macs is that you can use Fusion or Parallels
virtualization to turn a "Dual Boot" system into a "dual concurrent"
system running under Mac OS/X.
Mac's BSD kernel isn't quite as efficient with memory allocation and
garbage collection as Linux is, but it's still MUCH better than
Windows. I've worked with Several Mac users who were on engagements
with me and would use virtualized Windows for things like Visio and
Project, and virtualized Linux for things like Rational Application
Developer and other Java things that weren't supported on "Native" OS/
Mac and 64 bit Linux are both great "Hosting" operating systems for
secondary machines because they are very fast. The other nice thing
is that when you use external USB drives for the VMs, you can have
several different VMs based on your needs.
Windows XP and Vista suffer horribly from garbage collection problems
and switching frequently between VM and native or even between VMs
tends to throw Vista into garbage collection almost immediately.
One of the things Linux does to reduce the time spent garbage
collection is to allocate "large blocks" or "small blocks" from the
kernel. The allocation of amounts smaller than the blocks is done by
the application library, and often, applications malloc increments of
the block size so that even the application doesn't need to wait for
Since the blocks are merely memory mapped, there is no need to do
anything other than memory map. The L1 and L2 caches tend to make
even heavily fragmented memory look as if it's contiguous.
Keep in mind that when Linus was first writing Linux, he was
fascinated by the MMU on the 80386 processor. He wanted to really
exploit the daylights out of it, and it became a crucial and unique
element of his design. It's one of the reasons that the Linux kernel
is radically different, internally, from the older Minux, HURD, MACH/
BSD, and AT&T kernels.
> Linus runs Linux on Apple hardware (and see article below).
> Torvalds rejects one-size-fits-all Linux
> ,----[ Quote ]
> | Linus Torvalds has rejected the argument that Linux developers should pool
> | their resources behind a single distribution.
There is no point in trading one monopoly for another.
The main advantage of Linux is that it promotes aggressive competition
on a level playing field. The result is that all distributors have to
stay on top of the game AND adhere to common framework of standards to
stay in the game, and all of them can pick up nice tidy profits in the