[ale] Linux and Sun

stephen stephen at phynp6.phy-astr.gsu.edu
Fri Nov 30 11:07:53 EST 2001


On Fri, Nov 30, 2001 at 08:41:25AM -0700, Chris Ricker wrote:

<snip>

> 
> For the most part, I disagree.  Solaris is a disgusted bloated pig in
> comparison to Linux (and I say that as someone who admins and likes both
> ;-).  From a post I made to linux-kernel last time Dave M. wanted benchmarks
> on syscalls on the two:
> 
> <quote>
> 
> Script started on Tue Jun 19 17:49:30 2001
> [kaboom at thing2 linux]$ ./lat_proc fork
> Process fork+exit: 563.7778 microseconds
> [kaboom at thing2 linux]$ ./lat_proc fork
> Process fork+exit: 565.5556 microseconds
> [kaboom at thing2 linux]$ ./lat_proc fork
> Process fork+exit: 568.0000 microseconds
> [kaboom at thing2 linux]$ exit
> Script done on Tue Jun 19 17:49:46 2001
> 
> Script started on Tue 19 Jun 2001 05:51:38 PM MDT
> [kaboom at thing1 solaris]$ ./lat_proc fork
> Process fork+exit: 4249.5000 microseconds
> [kaboom at thing1 solaris]$ ./lat_proc fork
> Process fork+exit: 4212.5000 microseconds
> [kaboom at thing1 solaris]$ ./lat_proc fork
> Process fork+exit: 4241.0000 microseconds
> [kaboom at thing1 solaris]$ exit
> script done on Tue 19 Jun 2001 05:52:19 PM MDT
> 
> thing1 and thing2 are identical Sun Blade 100s.  thing1 is running Solaris 8
> (04/01 release), while thing2 is running 2.4.4 (Debian/unstable).
> 
> </quote>
> 
> You see the same thing with vfork(), etc.  Syscalls are literally an order 
> or two (depending on the call) of magnitude faster on Linux than on Solaris.
> 
> There are lots of good reasons why you might put Solaris on your Sun box 
> instead of Linux, but speed is not one of them.

This may be true in kernelland, but I don't think you can build 64-bit
applications to run under sparc64.  I once did a simple PVM application
that performed matrix multiplication and ran it on 3 nodes; an ultrasparc5
(with a 370mhz processor), a 200 mhz pentium, and a 450 mhz celeron.  Each
node received the a data set of the same size.  The ultrasparc5 only performed
marginally faster than the pentium 200, and was well out of proportion with
the celeron.  Now i never tested this against slowaris, but I believe that
without the ability to compile 64bit userland apps you're not going to use
the full potential of the ultrasparc processors.  But otherwise I love my
sparcstation as a developemnt box.

stephen

---
This message has been sent through the ALE general discussion list.
See http://www.ale.org/mailing-lists.shtml for more info. Problems should be 
sent to listmaster at ale dot org.






More information about the Ale mailing list