My old mate Shane said > I generally hate to argue with CC in public - much better done in person > where we can agree on the alcohol of choice before beginning.
True! And it's been a while between drinks. BTW, it's your shout. > With regard to Linux, I'm sure Ingo Molnar would take umbrage at the > above - significant advances have been made. As usual, just catching up > with our collective computing environment of choice, but getting there > none-the-less. But I was talking about UNIX. What I had in mind was SVR4.2 (and earlier) UNIX kernels. Partly because that was the last UNIX that I spent any time peering at kernel code and partly because it was the first "official" UNIX release that supported MP machines. My comment about dispatching contention was hardly controversial. Even UNIX bigots would probably accept it as a statement of fact for interactive usage cases even on a uniprocessor. It may well be better now (I don't know) but I did preface it with a "back in the day" in response to Paul Gilmartin's question about why people felt UNIX was limited to low utilization. So I'm not sure what we'd be arguing about, but since you're buying... have at it. > There is major (pre- ???) alpha code development of a workload > management paradigm under way - I happen to think it has lost it's way a > bit, but let's see what develops. But it'll be a while until it makes it > to the kernel mainline methinks. Probably so. Linux is getting to be quite a nice little OS. CC ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
