[snip] Sure its a valid comparison. The only thing a user cares about is the result. They might have a bit of tourist curiosity, but as long as they get their job done on time, they generally don't care HOW it gets done. So if it takes two or more redundant (but very cheap) boxes to do the job of a single (much more expensive) mainframe the user probably doesn't care. Moreover, the business may actually prefer the lower price, even if it comes at the cost of a bit more operational complexity. [snip]
We are wondering wide of my original point, but I hardly think you can replace a mainframe with two very cheap boxes. In our small shop here, we have 200-300 servers that need to be supported, and we *still* have a mainframe doing most of the bread-and-butter business. And I feel you are being disingenuous when you say "a bit more operational complexity". From my experience, adding truckloads of servers creates a lot more complexity, a lot more support staff, and a very convoluted, doubtful disaster recovery. I realize there is an industry wide push to recentralize servers, but I have not seen this done in practise yet in my neck of the woods. [snip] Actually, I -can- compare these things quite usefully. You appear to have an inherent preference for running all of these things on the same box at the same time. That's ok, but you should recognize it as a preference only. If the same thing can be accomplished more cost effectively, or more reliably, by some other approach, then who's to say the other approach is wrong? [snip] My original point was you can't compare a single-purpose box's reliability with that of a mainframe and get a fair comparison. This is a very narrow point, regardless of what the user sees and platform costs and whatnot. You're looking at a wider picture here. I have always believed that the idea of having many hundreds of servers is ridiculous, that processing should be centralized to give you better control, easier disaster recovery, and fewer support staff. But that's another kettle of fish. I actually agree with your interesting arguments up to a point, I'm not going to try to refute them. Any computing platform that is used by business to make money must be: reliable, easy to recover in a disaster, easy to maintain and debug, and deliver these things inexpensively with as few support staff as possible. In many cases it will make sense to run the business on a bunch of servers, but for large business I still feel the mainframe is the only platform that can deliver, despite its over inflated software costs. Perhaps in the next few years, if servers are consolidated sufficiently to a more managable level, they will be able to take over, but I don't think they've arrived yet. Certainly not around here anyway. If you are delivering mega-tons of grain over long distances to few destinations, it's better, and cheaper in the long run, to use a couple of mile-long freight trains, despite their high initial cost. If you are delivering thousands of envelopes to many destinations, it's better to have a fleet of chevettes. Each approach is correct for the product. What I inherantly disagree with is putting a few shovel fulls of grain in the trunks of thirty thousand chevettes, and saying that that's somehow better than using the train. Damn the cost, it just ain't right. Robin Murray Tel: (902) 453-7300 x4177 Cell: (902) 430-0637 ---------------------------------------------------------------------- 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
