You have absolutely zero (0) foresight.
Incorrect, I'm 100% practical and 100% skeptical.
> (try to get 1200 web operations per second on a Java VM based
> kernel using current processors, not gonna happen)
Personally I believe this is not because it cannot be done --- but
rather because no one knows how to do it, and those who might
aren't.
When people go into this "OS of the year 2000" routine, and say let's
make our stuff "object oriented" etc. whatever crapola they can make
sound like good marketing amunition, the first few things that comes
to my mind is:
1) The technology doesn't exist yet in any form that you'd actually
ask someone to run their main systems on, this means researcher
toys do not count
2) We have a proven technology and design scheme, which we are using
on real systems, and I question anyone who wants to jump out into
a new frying pan when we (IMHO) still haven't perfected monolithic
kernels (yet)
Therefore until someone shows some proven real working examples of the
ideas, let's continue to be practical people making systems which are
useful and continue to squeeze every last cycle out of the kernel that
we can. The pure researchers will let us know when they have some new
scheme which we can investigate and thus evaluate as suitable or not.
---------------------------------------------////
Yow! 11.26 MB/s remote host TCP bandwidth & ////
199 usec remote TCP latency over 100Mb/s ////
ethernet. Beat that! ////
-----------------------------------------////__________ o
David S. Miller, davem@caip.rutgers.edu /_____________/ / // /_/ ><