You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Mike Abbott <mj...@trudge.engr.sgi.com> on 2000/08/24 23:20:20 UTC

AAP patch for 2.0a6 available

The Accelerating Apache Project announces the availability of its
enhancements for Apache/2.0a6.
	http://oss.sgi.com/projects/apache/

We also are pleased to report that on a dual Pentium III Linux 2.4
system our state-threaded MPM is 25% faster than the dexter MPM, which
uses a similar process/thread architecture with pthreads.
-- 
Michael J. Abbott        mja@sgi.com        http://reality.sgi.com/mja/

Re: AAP patch for 2.0a6 available

Posted by Daniel Veillard <Da...@w3.org>.
On Sun, Aug 27, 2000 at 05:01:03AM -0700, Manoj Kasichainula wrote:
> On Thu, Aug 24, 2000 at 02:20:20PM -0700, Mike Abbott wrote:
> > The Accelerating Apache Project announces the availability of its
> > enhancements for Apache/2.0a6.
> > 	http://oss.sgi.com/projects/apache/
> > 
> > We also are pleased to report that on a dual Pentium III Linux 2.4
> > system our state-threaded MPM is 25% faster than the dexter MPM, which
> > uses a similar process/thread architecture with pthreads.
> 
> This is pretty cool. I've been wanting to add better support for
> something like Pth to Dexter.

  Just read http://oss.sgi.com/projects/apache/stm.html
and http://oss.sgi.com/projects/state-threads/docs/st.html

 This sounds like the perfect combination at least on the Unices
platforms: multiplexing processors using multiple processes and
multiplexing connections within processes using user level state
based threads. Sounds a lot like the best one can do with the little
caveat that one thread cannot be serviced by another process (or
kernel thread) like one can achieve for example using Mach C
threads (or other similar thread library where user threads
are scheduled on top of a more limited set of kernel threads).
On the other hand having separate processes is so important for
the overall reliability that it overrides IMHO the small performance
penalty of the state thread approach.

   looks really good to me,

Daniel

-- 
Daniel.Veillard@w3.org | W3C, INRIA Rhone-Alpes  | Today's Bookmarks :
Tel : +33 476 615 257  | 655, avenue de l'Europe | Linux XML libxml WWW
Fax : +33 476 615 207  | 38330 Montbonnot FRANCE | Gnome rpm2html rpmfind
 http://www.w3.org/People/all#veillard%40w3.org  | RPM badminton Kaffe

Re: AAP patch for 2.0a6 available

Posted by Manoj Kasichainula <ma...@io.com>.
On Thu, Aug 24, 2000 at 02:20:20PM -0700, Mike Abbott wrote:
> The Accelerating Apache Project announces the availability of its
> enhancements for Apache/2.0a6.
> 	http://oss.sgi.com/projects/apache/
> 
> We also are pleased to report that on a dual Pentium III Linux 2.4
> system our state-threaded MPM is 25% faster than the dexter MPM, which
> uses a similar process/thread architecture with pthreads.

This is pretty cool. I've been wanting to add better support for
something like Pth to Dexter.

I haven't had a chance to pore through the code (and I had trouble
getting to the patch when I tried anyway). Is this a modification of
Dexter, or a whole new MPM? How much of this is implemented behind the
scenes in the state threading library and how much is exposed in the
Apache layer?

It looks like it wouldn't be hard to write a pthreadish wrapper for
this. Then, Apache could have a
--with-threads=(pthread|pth|statethreads) configure option.

Good stuff.