Hehe... yeah, the geniuses in our Capacity Planning group ran into that problem as well. For a week afterward, they were busy wandering from VP's office to VP's office, trying to explain to everyone why they'd thought it was a good idea at the time...
Given 1,000 virtual Linux images running on a single mainframe LPAR using VM:
- how many virtual power supplies are going to need to be replaced?
- how many square feet of real estate are going to be used up by their virtual racks?
- how many miles of virtual cable and/or fiber are going to be laid underneath the clean room floor?
- how much does an idle virtual server cost to operate during off-peak hours? (OK, it's a trick question. When you can dial up new images on-demand, you need never run more virtual servers than are absolutely required.)
- when capacity planning finally admits their estimates were 20% below actual usage, how much will it cost (in both time and money) to dial up another 20 virtual servers to meet the workload?
- how many virtual servers will receive an automatic 'upgrade' when the host box gets a performance boost?
It's funny how people say "Linux is great for legacy hardware" when talking about their $500 486sx, but not for a $500k s370.
This sounds like a pretty old box. Most of the newer z/Series boxen are configured with additional CPUs that can be tapped for unusually high peaks in processing, on a pay-as-you-go basis.
For example, you can rent a box with 12+4 CPUs. Ninety-nine percent of the time, the 12 CPU you paid for are good enough. However, at the end of year, you tend to run at 120% capacity. All you have to do is call up IBM to activate the 4 additional CPUs and increase your capacity. Come January, you call IBM back up and they turn off the spare CPUs. You only pay for a 16 CPU box for the month you use it.
As for *nix on the mainframe, IBM is doing some pretty innovative licensing schemes. For example, you can identify several of the CPU in your box as Linux-only. Since the price of most mainframe software is based on how many MIPS and/or CPUs your box has, this can help reduce costs for both OSes.
Yeah, but I can use Gnutella to listen to every track on those albums, decide which ones I want to buy, and then download the 10 songs I really want, saving me ~$40.
Or,I could buy all 3 albums @ $9.99 a pop, saving me only $20.
According to your logic, if the gamble doesn't pay off in 5 years, would they get to write it off as a loss then? Or are you saying write-offs shouldn't be allowed at all? If so, wouldn't this make new venture spending even less appealing for businesses?
Wouldn't this just stifle innovation in the marketplace?
It's pitched at the corporate market. They're really just talking about outsourcing.
Why buy hardware, software, personnel, etc., when you can hire us (IBM) to do it for you. Rather than a set contract, we'll just bill you as you go along.
How different really is a browser-based client-server application from a mainframe-3270 terminal app?
Try reading the actual report (which wasn't sponsored by Linux Today, btw). There's a link to a PDF of it on the same page as the article. Check under 'Methodology' for the how they analyzed TCO.
Archer screws up. A day later, all life on Earth is destroyed. Archer and Future Guy reappear in the future. Earth is exactly the way Archer left it, which isn't suprising given that no one has stepped foot on the planet since it was destroyed.
The only mystery here is why wasn't his apartment covered in 900 years of dirt and dust?
Take a look at the Skycat line of airships. They're practically marketed as long-range troop movers, capable of moving up to 1,000 tons faster and cheaper than conventional land and sea methods.
In 25 years, we will all be using the same PCs we are today. They'll be smaller, and software will be better, but we really won't need any more power than we have now.
Isn't that the same thing IBM said, oh, 25 years ago?
This should just about kill MMORPG's (like UO, EQ, AC, AO, DaoC) customer base. Hell, these people are already paying a premium just to logon.
And how many people are still going to host Unreal, Quake, and Doom servers when their bandwidth is on the line?
For that matter, what about VPNs? If tunnelling in to work through my cable modem is going to cut into my gametime, they'd damn well better be prepared to help foot the bill!
...and then, when Store A gets bought out by Mega-Corp Inc., all the new owners will have to do is send out a 'our terms of service have changed' e-mail. Those that don't read the new terms, much less follow the 10 links to the opt-out screen, get turned into revenue.
We have no leverage to keep the terms from changing. Resistance is futile.
So, when you run out of floor space, do you shut down your shop for the week(s) it takes to detect failed modules, tear the complex apart, remove/replace them, and rebuild the collective?
Does IBM somehow believe that floorspace is infinite and free, while regular maintenance is cumbersome and expensive?
That's SuSE's 7.0 distro, using the 2.2 kernel (and associated relevant packages). The stable 2.4 kernel release doesn't appear anywhere, although I did happen to track down a beta copy from their dev site. No.ISOs, though.
SuSE's s390 sales brochure is a little sneaky; it doesn't really make it clear that they're only charging for service and support. Nor is there any good explanation for why they adopted an MVS 'per CPU' licensing scheme, especially when the same sheet shows no price difference for single- multiple-CPU PCs.
If you ask me, it sounds like they're trying to make the most out of the ignorance of mainframe managers.
Since you brought up AIX...
OS/390 includes a copy of an AIX-compliant Unix VM already. We've been running Notes servers under it for several years now. IMHO, foregoing the existing Unix implementation on the mainframe for Linux may just be a step backward.
Slashdot Mirror
Hehe... yeah, the geniuses in our Capacity Planning group ran into that problem as well. For a week afterward, they were busy wandering from VP's office to VP's office, trying to explain to everyone why they'd thought it was a good idea at the time...
Given 1,000 virtual Linux images running on a single mainframe LPAR using VM:
- how many virtual power supplies are going to need to be replaced?
- how many square feet of real estate are going to be used up by their virtual racks?
- how many miles of virtual cable and/or fiber are going to be laid underneath the clean room floor?
- how much does an idle virtual server cost to operate during off-peak hours? (OK, it's a trick question. When you can dial up new images on-demand, you need never run more virtual servers than are absolutely required.)
- when capacity planning finally admits their estimates were 20% below actual usage, how much will it cost (in both time and money) to dial up another 20 virtual servers to meet the workload?
- how many virtual servers will receive an automatic 'upgrade' when the host box gets a performance boost?
It's funny how people say "Linux is great for legacy hardware" when talking about their $500 486sx, but not for a $500k s370.
This sounds like a pretty old box. Most of the newer z/Series boxen are configured with additional CPUs that can be tapped for unusually high peaks in processing, on a pay-as-you-go basis.
For example, you can rent a box with 12+4 CPUs. Ninety-nine percent of the time, the 12 CPU you paid for are good enough. However, at the end of year, you tend to run at 120% capacity. All you have to do is call up IBM to activate the 4 additional CPUs and increase your capacity. Come January, you call IBM back up and they turn off the spare CPUs. You only pay for a 16 CPU box for the month you use it.
As for *nix on the mainframe, IBM is doing some pretty innovative licensing schemes. For example, you can identify several of the CPU in your box as Linux-only. Since the price of most mainframe software is based on how many MIPS and/or CPUs your box has, this can help reduce costs for both OSes.
Yeah, but I can use Gnutella to listen to every track on those albums, decide which ones I want to buy, and then download the 10 songs I really want, saving me ~$40.
Or,I could buy all 3 albums @ $9.99 a pop, saving me only $20.
According to your logic, if the gamble doesn't pay off in 5 years, would they get to write it off as a loss then? Or are you saying write-offs shouldn't be allowed at all? If so, wouldn't this make new venture spending even less appealing for businesses?
Wouldn't this just stifle innovation in the marketplace?
It's pitched at the corporate market. They're really just talking about outsourcing.
Why buy hardware, software, personnel, etc., when you can hire us (IBM) to do it for you. Rather than a set contract, we'll just bill you as you go along.
How different really is a browser-based client-server application from a mainframe-3270 terminal app?
Try reading the actual report (which wasn't sponsored by Linux Today, btw). There's a link to a PDF of it on the same page as the article. Check under 'Methodology' for the how they analyzed TCO.
Archer screws up.
A day later, all life on Earth is destroyed.
Archer and Future Guy reappear in the future. Earth is exactly the way Archer left it, which isn't suprising given that no one has stepped foot on the planet since it was destroyed.
The only mystery here is why wasn't his apartment covered in 900 years of dirt and dust?
...after they assimilated 10 years worth of Starsearch and American Idol runner-ups.
Resistance is futile.
I know this is off-topic, but your sig is inaccurate. The actual quote is 'They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.' Quite a different sentiment. Sorry to hijack your thread.
Take a look at the Skycat line of airships. They're practically marketed as long-range troop movers, capable of moving up to 1,000 tons faster and cheaper than conventional land and sea methods.
In 25 years, we will all be using the same PCs we are today. They'll be smaller, and software will be better, but we really won't need any more power than we have now.
Isn't that the same thing IBM said, oh, 25 years ago?
There's a big difference there....
Mod this up!
This should just about kill MMORPG's (like UO, EQ, AC, AO, DaoC) customer base. Hell, these people are already paying a premium just to logon.
And how many people are still going to host Unreal, Quake, and Doom servers when their bandwidth is on the line?
For that matter, what about VPNs? If tunnelling in to work through my cable modem is going to cut into my gametime, they'd damn well better be prepared to help foot the bill!
...and then, when Store A gets bought out by Mega-Corp Inc., all the new owners will have to do is send out a 'our terms of service have changed' e-mail. Those that don't read the new terms, much less follow the 10 links to the opt-out screen, get turned into revenue.
We have no leverage to keep the terms from changing. Resistance is futile.
So, when you run out of floor space, do you shut down your shop for the week(s) it takes to detect failed modules, tear the complex apart, remove/replace them, and rebuild the collective?
Does IBM somehow believe that floorspace is infinite and free, while regular maintenance is cumbersome and expensive?
In a response, JFK International Airport had this to say:
"The Doppler-5000 is a great radar system... but would I use it to forecast stock performances right now? No."
That's SuSE's 7.0 distro, using the 2.2 kernel (and associated relevant packages). The stable 2.4 kernel release doesn't appear anywhere, although I did happen to track down a beta copy from their dev site. No .ISOs, though.
SuSE's s390 sales brochure is a little sneaky; it doesn't really make it clear that they're only charging for service and support. Nor is there any good explanation for why they adopted an MVS 'per CPU' licensing scheme, especially when the same sheet shows no price difference for single- multiple-CPU PCs.
If you ask me, it sounds like they're trying to make the most out of the ignorance of mainframe managers.
On behalf of os/390 system programmers everywhere, I'd like to say, "we love ya too". Now, get outta here, ya big lug...>sniff,sniff...
We've been using CMOS processor complexes for the better part of the last 2 or 3 years. Is this really new, or am I missing something?
And worshiping a god that is not one, not two, but three, three, THREE beings in one shows your grasp of reality how exactly???
Aside from the intro, this MP3 is a rip of the sound track from the Summoner D&D-goof movie.
You forgot to mention how D&D also turns us all into homosexual cross-dressers that like to say "niep!" and worship shrubbery....
Since you brought up AIX... OS/390 includes a copy of an AIX-compliant Unix VM already. We've been running Notes servers under it for several years now. IMHO, foregoing the existing Unix implementation on the mainframe for Linux may just be a step backward.