Slashdot Mirror
IBM Plans Collaboration On Power Architecture
TheInternet writes "According to CNET News, IBM has made a series of announcements regarding the opening-up of the Power chip architecture. The story lacks technical details, but apparently, IBM is going to divulge more information about Power/PowerPC, and expects collaboration from the industry on the future of the chip. Nick Donofrio is quoted as saying: 'We will free electronics manufacturers from the limitations of proprietary microprocessor architectures', and Red Hat and Sony are two companies listed as taking part. Power5 was also shown, as was the Blue Gene/L supercomputer, using 32 500MHz processors to achieve 128 gigaflops."
Another interesting .
link at the Inquirer
Seems IBM is courting third party mobo makers to make PowerPC boards.
Their emracing linux and opening up their hardware platform. Sound Like
their getting their troops in line for THE desktop battle.
I, for one, would love to be running Debian Linux on a ATX PowerPC board. Of
course, they would have to sell enough of them to get the price down.
Good luck to 'em.
So, does this mean we might see some good PowerPC emulators coming out? I remember reading that one of the main problems with writing them was the fact that the chip was quite closed, so one essentially had to reverse-engineer the entire instruction set.
From Tom's Hardware: IBM's processor plans: Build your own microchip.
Do you know why the road less traveled by is littered with the bones of the unwary?
Anyone else think this on the first read?
Does this sound like a possible good contender for a general purpose replacement for x86 as an "Open CPU" that would work well with F/OSS apps? One that can't be tied down with DRM in such a way that only large megacorps (I won't name them, except to call them "Microsoft" and "Intel". Oh hell I named them whoops) can end up defining what may or may not run?
I've read the general slashdot crowd clamoring for something like this that's free from central control. Does this look likely to you? Would it be a benefit if it did come about?
This fits with IBM's vision for spreading the 970. There's two groups: "Pervasive" and "Deep." IBM uses "pervasive" to describe a wide range wired and wireless devices powered by the 970 chips, (i.e. p2p sharing of naked petrified natalie portman pictures). "Deep" computing describes IBM's high performance technical computing products, like Blue Gene.
Opening the architecture swings the door for pervasive market penetration, indeed.
The theory of relativity doesn't work right in Arkansas.
128 Gigaflops? April's Fools!
(Hey, it's started already, just look at that pigeon story).
Cyde Weys Musings - Scrutinizing the inscrutable
There are sound business reasons for all the big-headline stuff IBM's been doing lately, but I think they're doing it because it's the biggest, best, and cheapest PR anyone's ever heard of.
Sure they're positioning POWER/PowerPC as the only architechture that can challenge x86 in a meaningful way. Sure if they release enough of the firmware and stuff it'll probably be better for some open source stuff than x86 is. Sure they're a services company and this will put them on the back end of even more stuff.
But honestly. Everyone loves them. That's what they really need if they want to entrench themselves everywhere.
I rarely criticize things I don't care about.
I, for one, welcome our OpenPowerPC overlords.
No, seriously, I think this is a great step. When we get to control the functionality and content of our silicon, and contribute to the specs, I think a LOT of creative people will come forward and throw out some truly awe-inspiring ideas. Look what happened with Linux, *BSD, countless GNU projects. The list goes on, people. I think this could be a stepping stone towards getting some really new chip technology on a roll.
Lets just hope this is a sincere effort on IBM's behalf.
Umm, the instruction set is documented and everything. There's this PDF you can download.
The problems with emulators have to do with RISC vs CISC differences and register-rich vs register-poor architectures. I have to go, so I'm not going to go into the details here, but the general idea is this: for the specific case of emulation, it's easier to write an emulator if your host architecture is more RISCy than your emulated architecture, and it's easier to write an emulator if your host architecture has more registers than your emulated architecture.
The PowerPC has a very cluttered instruction set, but it still basically follows RISC as a philosophy-- you're still in a situation where instructions from other architectures have mostly instructions that can be broken down efficiently into a series of PowerPC instructions. Which means that efficiently assembling series' of PowerPC instructions into single instructions while emulating on more CISCy platforms is kind of hard. The PowerPC also has a whole lot of registers, and they're all general purpose so you can't play neat optimization tricks as easily as you can when emulating the Intel x86. Meanwhile, the architecture you probably want to do this emulation on-- x86-- is shit for registers.
The PPC emulation problem has to do with unfortunate conflicts between design philosophies and emulation perverse cases more than anything else.
-- Super Ugly Ultraman
This is an important step, at least for the Power5. It's immensely complex, and I think feedback from collaborators such as OS people is important when they (IBM) ask themselves if a design decision makes sense. For example, SMT adds 24% to the die area for each core (see here). Compare that with Intel's HyperThreading, which adds little area but is still complicated to verify. Getting feedback and involving other groups can help determine if design decisions/features are worthwhile.
x86 and windows are established platforms and people and industry have put lot of time and effort in adopting them. why do you guys are interested in destroying this enormous value created with hardwork over time and replace with totally unproven vaporware system which only the geeks and use?
Even on April 1st, you don't play such cruel jokes.
IBM has seen how well the Open Source/Community model has worked for Linux. Now they believe that it will benefit the deployment of POWER derived technology.
The details are a little sketchy at this point, but Wladawsky-Berger basically said this is of the same magnitude as the decision to embrace Linux.
I think I heard the word "community" in almost every other sentence. I truly believe IBM "gets it" and is moving forward in bold direction. The people I talked to afterward were credible and excited.
There will be a longer story on ClusterWorld tomorrow. (sign up for three free issues of the magazine as well)
I saw the small "Blue Gene" system. Very cool both performance wise and thermally (32 CPUs in a table top box). I also saw the new Power blade server. Nice.
HPC for Primates. Read Cluster Monkey
That Blue Gene/L machine being shown is only a small part of the full machine they are building for LLNL. When its complete, IBM estimates that it will run at 360 TFLOPS, at a fraction of the size and power consumption of the current #1 supercomputer. Even if they miss the mark by 50% it represents a fairly significant leap in processing and power consumption. And hey, since it will only occupy 64 racks, you can just about fit one in your garage! (Nuclear reactor to power it not included...)
To handle this, Mac OS has generally included a "ROM" file which contained the part of Mac OS that had previously been stored on the ROM.
Mac OS and Mac OS X does do some checks before allowing itself to be installed on a machine, making sure it really is being installed on a Mac. I believe though that all it does ultimately is query OpenFirmware.
You are not alone. This is not normal. None of this is normal.
As you all know, the costs of developing new chip architectures is escalating. However, once designed, chips can be replicated at relatively low cost (at least by comptent fabs like IBMs). So, to maximize profits despite the high upfront costs, what is one to do?
Design a kick-ass chip, sign up a lot of partners to establish street credibility, maintain processor improvement momentum, deliver chips on time, then sell as many chips as possible, of course! AMD performed in some, but certainly not all of these aspects, hence their current standing in the chip industry. Don't even get me started on the slow train wreck called Motorola.
The power architecture was always meant to be flexible, ranging from the $10,000+ quad-core uber-chip Power5's on down... So it's only logical that we will find stripped-down versions of the Power5 architecture in everything from Apple Desktops to next-gen consoles from Sony and MS.
As I see it, this is a great PR step by IBM to get some mindshare from the growing Linux camp. When you combine the incredible performance, lower prices, etc. of the 970 architecture, folks like Intel will have to take notice sooner or later, particularly when it hits their most profitable processor lines. However, Apple may not be happy to face competition in a market segment that it has had to itself for now.
As for MS and their PowerPC line of NT or whatever, who cares. If they need to make the switch, they'll find a way. In the meantime, it's the Linux/Unix folks who'll benefit the most from no longer being squeezed between SPARC and XEON pricing.
The SPARC is already an open architecture processor. It's been that way for years. Sun was the big player behind it, and certainly the best known, but the SPARC design is the closest thing there is to an "open source cpu." There's even a non-Sun organization (SPARC International) they spun off to act as a steward for the standard.
SPARC processors are made by Sun, Texas Instruments, Hitachi, and others. There's a history of all the chips made on their web site.
Dunno why they're too blind to see that this would be as good an idea for Java.
Tired of FB/Google censorship? Visit UNCENSORED!