Messages in this thread | | | Date | Mon, 06 Sep 1999 05:49:56 +0000 | From | Steve Underwood <> | Subject | Re: > 15,000 Simultaneous Connections |
| |
Lincoln Dale wrote:
> At 05:27 AM 9/6/99 +0000, Steve Underwood wrote: > > > >I think a HOWTO on this subject might be in order :) > > > > > > i'm not sure this can be covered in a HOWTO document. > > > don't take this the wrong way, but if you have to ask how to code something > > > for efficiency in this manner, expecting to find samples in a HOWTO is > > > rather much a pipe dream. > > > >I think that is true of a genuine HOWTO. I do wish someone would produce a > >survey > >of the strategies used by major applications, though. It would make assessing > >existing solutions for their strengths and weaknesses so much easier. Its > >something large numbers of us do independantly, with consequent wasting of > >humanity's precious resources. I'm not volunteering, so I can't expect > >anyone else > >to either. It would be real useful, though. > > i agree that many people have individually gone over the same things over > and over, to find the best way to do something. > > however, it is a chicken-and-egg type problem -- > imho, you cannot get the knowledge to know what works best, until you've > experimented, and found the best way. > > a HOWTO document could only generalize. there's nothing stopping a > badly-written event-driven model or threaded model performing worse than > that of straight fork()/exec() type stuff.
I totally agree. That's why I specifically said a _survey_ and not an actual HOWTO. I think that a survey of currently used strategies would speed up our individual assessments of what is right for us. I suspect there are packages out there using interesting strategies that I have no time to study, or am not even aware of. Its exteremely time consuming looking through them. I would very much appreciate a survey that expresses as little opnion as possible. This could probably be said of many other areas of software design, too.
Steve
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |