• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Linux on a IBM dinosaur...

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
SO let me get this right. I've been reading a bit more on this stuff and it goes like this.

The term "dinosaur" refered to IBM mainframes doesn't just refer to them going out of style, It refers to the fact that it is a gigantic computer system controlled by a tiny brain.

Each disk controller does all the hardwork of shuffling information around, unlike the PC's were everything is proccessed thru the CPU..

The strength of the Mainframe is pure I/O power, not proccessing power. Kinda like Pc's in with 3-d graphics, if we relied on our main proccessor to run all the 3-d graphics, quake2 would be to much for our computers to handle with nice colors and high resolutions we expect. So we have dedicated hardware soly designed to proccess 3-d graphics, these graphics require a gigantic memory bus so that is all kept on the video card. Each itteration of directx and newer generation of video cards requires the cpu to do less and less work to produce excelent results.

Such is the way that the IBM mainframe functions. Each disk controller does all the grunt work, it's designed specificly for running the disks and such, so that the cpu is kept from having to worry about these things. That's how a mainframe can get all that work done.

So comparing a beowolf cluster, or a PC to a Mainframe is a apple and oranges comparision. Tasks at which a Mainframe would be suited at would geek a PC out, and a PC cluster can run rings around a Mainframe in terms of perfomance that we PC users are acustom to.

How accurate is that?
 
Originally posted by: drag
So comparing a beowolf cluster, or a PC to a Mainframe is a apple and oranges comparision. Tasks at which a Mainframe would be suited at would geek a PC out, and a PC cluster can run rings around a Mainframe in terms of perfomance that we PC users are acustom to.
Sounds about right to me, but then I don't know anything about this, either. 😉

Also, it's hard to stereotype mainframes, because some of them seem to be more CPU-power oriented. Some of SGI's Origin servers can have up to 512 CPUs (maybe more), but take "standard" SCSI disks and I/O cards (I believe they have PCI slots, plus some other high-performance interconnect slots as well). However, I'm not sure if they'd be considered a "mainframe" at all; they certainly are big enough, if size is what matters! 😱
 
drag: you're getting the idea. Starting to see how odd the entire design philosopy behind a mainframe and a unix-server are so different? A kernel can't be all things for all hardware, thats why linux and mainframes are such an odd mix. Eventually it'll take either a fork, a kernel thats so different that it mostly just shares the name, or a consession that performance isn't a priority.

jliechty: those don't count as frames. They're RISC chips running a *nix.
 
Originally posted by: Buddha Bart
jliechty: those don't count as frames. They're RISC chips running a *nix.
Ah, so we can only call them by the lowly term "server"? 🙁 😉

Thanks for the enlightenment, anyway. This thread is very informative. 🙂
 
Originally posted by: jliechty
Originally posted by: Buddha Bart
jliechty: those don't count as frames. They're RISC chips running a *nix.
Ah, so we can only call them by the lowly term "server"? 🙁 😉

Thanks for the enlightenment, anyway. This thread is very informative. 🙂

I believe SGIs setups are more like clusters than anything else. But, I'm probably wrong. 🙂
 
Originally posted by: jliechty
Originally posted by: Buddha Bart
jliechty: those don't count as frames. They're RISC chips running a *nix.
Ah, so we can only call them by the lowly term "server"? 🙁 😉

Thanks for the enlightenment, anyway. This thread is very informative. 🙂

You could always call it a "Bigass NUMA Supercomputer" or something if you want to 😉
 
Originally posted by: jliechty
Originally posted by: Sunner
That wasn't the part that confused me, I just assumed they went 360->370->380->390 with some minor changes in between those.
But then, we never got a 80186 or 68010 either 🙂
Uh, no... 😉

Heh, good one. They also had a Page on the Motorola 68010, too 🙂

By the way, Windows 2000 was considered to be NT 5.0. Windows XP/2003 was NT 5.1.
 
Originally posted by: ultimatebob
Originally posted by: jliechty
Originally posted by: Sunner
That wasn't the part that confused me, I just assumed they went 360->370->380->390 with some minor changes in between those.
But then, we never got a 80186 or 68010 either 🙂
Uh, no... 😉

Heh, good one. They also had a Page on the Motorola 68010, too 🙂

By the way, Windows 2000 was considered to be NT 5.0. Windows XP/2003 was NT 5.1.

Ok ok, no need to rub it in :|



😉

Learn something new every day, eh? 🙂
 
Back
Top