This has been discussed here and other places back when Google first came to prominence and claimed to be running their search engine on a bunch of PCs sitting on rickety tables.
It all comes back to the strengths of the mainframe, and IMHO they are not what Google needs or wants. It's not all that important to Google or its customers if one query here or there fails, or times out, or produces results inconsistent with the same query done a moment later. That Google's code might port quite easily to z boxes doesn't mean they'd be able to run what they have to, or that the price would be sensible. Just look at mainframe memory and disk prices: IBM dropped ram to "only" $US8000/GB or so on the z9; what do you suppose Google is paying at the moment for Intel memory? I know I pay around $100 for my home machines, and I imagine they do just a bit better than that... Similarly, much of the mainframe cost advantage is predicated on zillions of Windows boxes sitting on desktops with idle CPUs, whereas I imagine Google's Intels are kept very busy. It may well be easy to set up lots of Linux images on a z9, but they're not going to keep up with Linux images running one per Intel chip. I'd love to see it fly, but I'm not hopeful. Tony H. Tim Hare wrote: > If I were IBM, I would gather some of the smart Z/VM/Linux > folk, and I would go to places like Google, Yahoo!, or Six > Apar with some mainframes in a trailer already running a > hundred-or-so virtual Linux instances, for a week or three. > I'd let the company build their software on those Linux > images and test it out. ---------------------------------------------------------------------- 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
