Mary,
But, I admit that is more my bias and my main argument with Tony Viola. He (Tony) continues to believe there are
applications that require a legacy mainframe.
It's happening (big enterprise computing moving off of mainframes). I think when we started this whole "feud", which you called it in another post, there was tacit agreement that it would happen, but a difference in the timing of the prediction. As I recall, you had mainframes going off the cliff, or croaking relatively instantly like the dinosaurs (well, a million years or two out of hundreds of millions of years). I had it happening gradually. Six months ago, mainframes were going sideways in MIPS, a little down in $$ sales. Now, I think both are trending down. As Jean Gauthier points out, Sun is doing serious damage at the high end with their E10000 series. If they can do it, Intel (based) can with IA-64. Itanium is already running Solaris, and Win2000 is coming. Pretty sweeping statement, but why not? Of the things that mainframes had to themselves until recently (RAS, legacy 390 software and people afraid to move from it, CPU power, I/O bandwidth, what'd I miss?) Sun managed to either duplicate them or push them aside anyway (legacy). BTW, that Sun machine is one they got the rights to from SGI, which got it when they bought Cray Research. One of the best moves ever, by Sun.
So we're pretty much on the same page (another 90's phrase I hate) just timing a bit off. The door is definitely open, for those that want to get into big iron.
Tony |
