On Sun, 18 Dec 2005 14:54:15 +0100, R.S. <[EMAIL PROTECTED]> wrote: >John S. Giltner, Jr. wrote: >[...] >> Mostly Linux on the MF was so that you could migrate what was on 10, 20, >> 200 Intel boxes that were sitting there 90% idle and put on a single box >> and share CPU and Memeory resources. > >IMHO this is marketing mantra from IBM. >1. Sharing memory resource is not true. Memory can be easily >reconfigured (taken from system A to system B), but, assuming 24x7 >availability it is big advantage.
1a) Sharing memory resource IS true if you are using z/VM, as the site where I'm backup VM-guy is doing with its mainframe Linux workloads. They have 24x7 (well, closer to 24x6.98 with periodic scheduled outages for non- VM reasons). If it was important to the business to be fully up at 5am on every Sunday morning they could achieve 24x7 but it is more important to the business that it spend that money elsewhere. >2. Usually all the servers have workload peak in the same time. They >usually form single information system. 2a) Not true. Our Linux workload is sufficiently diverse to allow the shared workload without worry or bother. >3. Even If sharing resources could make some savings in term of >megabutes or CPU cycles, it doesn't necessarily mean any savings in >costs. CPU cycles and megabytes in PCs are much cheaper. Maybe cycles >are hard to compare, but memory price is simple to compare - megabyte >has the same meaning. 3a) IFL CPU cycles are less expensive than non-IFL mainframe CPU cycles, but that's not where any business site spends the majority of its money. It is the PEOPLE COST that is the biggest issue and being able to spawn a new Linux instance from VM - automatically, as demand rises & falls, is where VM shines. Well, that and the expense of the power (electricity & cooling) needed for the virtual vs. real environments. A few racks of servers are roughly the same power & cooling as a modern mainframe. A LOT of racks of servers require substantially more (1) cooling, (2) electricity and (3) people to roam through replacing failed parts than a mainframe. Well designed VM scripts can provision a new Linux instance far, far faster and much cheaper than a PC. >4. There is also not mentione advantage of Linux on z/machine: hardware >reliability and availability. It is hardly comparable, the only thing in >replace of RAS could be hot reserve or clustering. 4a) Reserve or clustering is a workable approach to counter the mainframe's RAS - certainly that's how the AIX crowd looks at it, but you still have the power/cooling/people costs that more than offset the "savings". Spintronics might offer a substantial improvement for PCs but that's more than 5 years away (probably more like 12) and it will impact the mainframe equivalently so the balance won't shift. Take a look around your computer site today and remember what it was like 30 years ago. You have far fewer mainframe people (for the size of your mainframe) and probably more total people on your site, with most of the body count against either (a) PC support or (b) administrative overhead. Most sites I've visited have fewer mainframe people even without considering the growth in the processors -- it is less people-intensive than it used to be -- but more "junior executives" than the old days. Dump the Jr. Execs if you want to save some serious money. -- Tom Schmidt Madison, WI ---------------------------------------------------------------------- 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
