The new Xserve cluster, says Dr. Shinpaugh, is “a very beautiful-looking machine, a combination of black racks and brushed aluminum, finished very nicely. Then you have the nice blue blinky lights. A lot of people who come by like to see blinky lights so they can say, ‘Oh, there is something going on.’:)
Tuesday, March 22, 2005
Best supercomputing quote ever
Apple has put up a new description of Virginia Tech's System X supercomputer. In it, Dr. Kevin Shinpaugh, Director of Research and Cluster Computing, has this to say about System X:
Sunday, March 20, 2005
IBM's 970MP: Top 10 things to know about it
Even though there is now all this 970MP hoopla, it seems that many people don't really have an idea of what it really is. Rumours have been all over the map about it recently, but to the best of our knowledge it seems that the Think Secret article that broke the news about the 970MP a year ago is probably the most accurate.
In this article I will outline the top ten things (in no particular order) we should know about this chip:
1) As the name would suggest, it is a 970 series chip, but dual-core. In other words, it is a dual-core chip that shares the POWER4 heritage of the 970FX. It is not a POWER5 derivative, nor is it a derivative of Cell's PPE.
2) It will not support simultaneous multithreading. If it had been a POWER5 derivative, this feature might have appeared, since SMT is a feature of the POWER5.
3) It will likely not have an integrated memory controller. An integrated memory controller would have been a bonus, but the information available makes no mention of an IMC. (The POWER5 has an integrated memory controller, but the POWER4 does not.)
4) It will likely add support for partitioning. This is a feature absent on the 970FX, but which is present in the POWER4. Don't ask me to explain it though, since I'm a bit fuzzy on the concept myself, especially in terms of how it would fit in with Apple's products.
5) Think Secret said it will have 1 MB L2 cache per core. This makes sense as the cache upgrade is a natural evolution for the 970 series chips. At higher clock speeds, it will be more necessary too. There is no information available about a shared cache, but it's likely the caches will be independent, and not a combined shared 2 MB cache for both cores.
There is still no official confirmation of the 970GX (although one article at IBM, but not written by IBM, did reference this chip). The 970GX is the single-core variant 970MP (with the increased L2 cache), and such a chip would be perfect for the iMac and the PowerBook, and possibly also for a low end single-CPU Power Mac. One wonders if the 970GX could simply be a 970MP where one of the cores is defective and has been deactivated (or even completely removed). Recycling partially defective 970MP chips as single-core chips would reduce overall costs.
6) The 970MP is fabricated on IBM's 90 nm CMOS SOI10K process, and has a die size of 153.8 mm² . For comparison, Intel's dual-core Pentium D Smithfield will be 206 mm², or approximately one-third larger. The current G5 970FX is 66 mm².
7) We don't know about on-chip communication speeds between the two cores of the 970MP. Some dual-core chips to be released this year depend on the FSB for the cores to communicate with each other, but this is not ideal. Preferably the 970MP would have higher on-chip communication speeds between the two cores, but we have no information either way.
[Update 2005-03-21] There is however, this statement from the original 970MP rumour: "The IBM PowerPC 970MP incorporates two complete microprocessors on a single chip, along with some common logic to connect these microprocessors to a system." This doesn't really answer our question, but it certainly does not confirm that the two cores can communicate with each other at higher than FSB speeds. [/Update]
8) We don't know the 970MP's clock speed, but we can make some guesses. Right now Apple states that the 2.3 GHz G5 in the Xserve runs at a maximum of 55 Watts. That is an impressively low amount of power consumption for a chip with its level of performance. (For comparison, the 90 nm Intel Pentium 4 Prescott chips run at around 130 Watts.) There are also hints that the 2.4 GHz G5 might run at 60 Watts. Momentum's 970FX evaluation motherboard is listed to support G5s up to 2.4 GHz, and in their board block diagram it lists a CPU power specification of 60 Watts. We don't know the power specification of the 2.5 GHz G5, but we could guess that it would be in the range of 65ish Watts.
Thus, we can expect the dual-core G5 970MP to debut with a clock speed of at least 2.5 GHz. One can even hope that IBM has got things working well enough to make a 2.8-3.0 GHz dual-core release possible. The information provided by Think Secret previously was that a 3 GHz dual-core (and a 3.0 GHz single-core) release was the goal, and we already have Steve Jobs' 3 GHz promise back from WWDC 2003.
9) We don't know what the bus speed will be in Macs. Think Secret suggests a 3:1 divider, which would mean a 1 GHz bus for a 3 GHz chip. However, it would also mean something like a 733 MHz bus for say a low end 2.2 GHz dual-core Power Mac. This is slower than the current low end 1.8 GHz dual Power Mac, but for most usage it may not be a major issue, since memory speed would not be significantly higher with DDR400, even with dual-channel memory. (As you may suspect, the memory type is not dependent on the 970MP at all. Like the choice of AGP vs. PCIe, the choice of memory used in the next Power Mac is up to Apple. So why not DDR2? It's possible the next Power Mac could get DDR2 support, but DDR2-533 offers little overall performance advantage over DDR400, and DDR2-667 is still quite expensive.)
10) The 970MP is designed to support up to a 4-way configuration, which would support the previous (dual dual-core) quad Power Mac rumours.
Interesting times ahead for Apple...
[Update 2005-03-21]
How could I forget? Let's just add one more.
11) IBM's leaked 970MP article references the Momentum G5 970FX evaluation board mentioned above (#8), and states that it was used to test the 970MP's thermal diode.
In this article I will outline the top ten things (in no particular order) we should know about this chip:
1) As the name would suggest, it is a 970 series chip, but dual-core. In other words, it is a dual-core chip that shares the POWER4 heritage of the 970FX. It is not a POWER5 derivative, nor is it a derivative of Cell's PPE.
2) It will not support simultaneous multithreading. If it had been a POWER5 derivative, this feature might have appeared, since SMT is a feature of the POWER5.
3) It will likely not have an integrated memory controller. An integrated memory controller would have been a bonus, but the information available makes no mention of an IMC. (The POWER5 has an integrated memory controller, but the POWER4 does not.)
4) It will likely add support for partitioning. This is a feature absent on the 970FX, but which is present in the POWER4. Don't ask me to explain it though, since I'm a bit fuzzy on the concept myself, especially in terms of how it would fit in with Apple's products.
5) Think Secret said it will have 1 MB L2 cache per core. This makes sense as the cache upgrade is a natural evolution for the 970 series chips. At higher clock speeds, it will be more necessary too. There is no information available about a shared cache, but it's likely the caches will be independent, and not a combined shared 2 MB cache for both cores.
There is still no official confirmation of the 970GX (although one article at IBM, but not written by IBM, did reference this chip). The 970GX is the single-core variant 970MP (with the increased L2 cache), and such a chip would be perfect for the iMac and the PowerBook, and possibly also for a low end single-CPU Power Mac. One wonders if the 970GX could simply be a 970MP where one of the cores is defective and has been deactivated (or even completely removed). Recycling partially defective 970MP chips as single-core chips would reduce overall costs.
6) The 970MP is fabricated on IBM's 90 nm CMOS SOI10K process, and has a die size of 153.8 mm² . For comparison, Intel's dual-core Pentium D Smithfield will be 206 mm², or approximately one-third larger. The current G5 970FX is 66 mm².
7) We don't know about on-chip communication speeds between the two cores of the 970MP. Some dual-core chips to be released this year depend on the FSB for the cores to communicate with each other, but this is not ideal. Preferably the 970MP would have higher on-chip communication speeds between the two cores, but we have no information either way.
[Update 2005-03-21] There is however, this statement from the original 970MP rumour: "The IBM PowerPC 970MP incorporates two complete microprocessors on a single chip, along with some common logic to connect these microprocessors to a system." This doesn't really answer our question, but it certainly does not confirm that the two cores can communicate with each other at higher than FSB speeds. [/Update]
8) We don't know the 970MP's clock speed, but we can make some guesses. Right now Apple states that the 2.3 GHz G5 in the Xserve runs at a maximum of 55 Watts. That is an impressively low amount of power consumption for a chip with its level of performance. (For comparison, the 90 nm Intel Pentium 4 Prescott chips run at around 130 Watts.) There are also hints that the 2.4 GHz G5 might run at 60 Watts. Momentum's 970FX evaluation motherboard is listed to support G5s up to 2.4 GHz, and in their board block diagram it lists a CPU power specification of 60 Watts. We don't know the power specification of the 2.5 GHz G5, but we could guess that it would be in the range of 65ish Watts.
Thus, we can expect the dual-core G5 970MP to debut with a clock speed of at least 2.5 GHz. One can even hope that IBM has got things working well enough to make a 2.8-3.0 GHz dual-core release possible. The information provided by Think Secret previously was that a 3 GHz dual-core (and a 3.0 GHz single-core) release was the goal, and we already have Steve Jobs' 3 GHz promise back from WWDC 2003.
9) We don't know what the bus speed will be in Macs. Think Secret suggests a 3:1 divider, which would mean a 1 GHz bus for a 3 GHz chip. However, it would also mean something like a 733 MHz bus for say a low end 2.2 GHz dual-core Power Mac. This is slower than the current low end 1.8 GHz dual Power Mac, but for most usage it may not be a major issue, since memory speed would not be significantly higher with DDR400, even with dual-channel memory. (As you may suspect, the memory type is not dependent on the 970MP at all. Like the choice of AGP vs. PCIe, the choice of memory used in the next Power Mac is up to Apple. So why not DDR2? It's possible the next Power Mac could get DDR2 support, but DDR2-533 offers little overall performance advantage over DDR400, and DDR2-667 is still quite expensive.)
10) The 970MP is designed to support up to a 4-way configuration, which would support the previous (dual dual-core) quad Power Mac rumours.
Interesting times ahead for Apple...
[Update 2005-03-21]
How could I forget? Let's just add one more.
11) IBM's leaked 970MP article references the Momentum G5 970FX evaluation board mentioned above (#8), and states that it was used to test the 970MP's thermal diode.
Using the PowerPC970FX evaluation board as an example, a Cypress Semiconductor PSOCTM (PSOC), P/N CY8C27243, is chosen to sample the thermal diode’s voltage, convert the voltage to a numerical value, and send the information to the service processor for disposition. The PSOC contains configurable analog and digital blocks to create the needed A/D converter and other analog functions.This suggests that the 970MP might be pin compatible with the 970FX, although we do not know for sure.
Tuesday, March 15, 2005
CHUD 4.1.1: References to quad CPUs and 970MP removed
CHUD 4.1.0 provided many clues to future dual-core and quad (dual dual-core) Macs, with its references to "970MP" and "Quad", as well as MONster.app's obvious support of 4 cores:
However, with the new 4.1.1 version, all the above clues have been removed. MONster.app now just shows a single CPU for my Cube, with no greyed-out CPU 2, CPU 3, and CPU 4. Interestingly however, the font size in 4.1.1 is the same as in 4.1.0, which means there is sufficient space to display CPUs 2-4, if the cores are present:
It's obvious now that 4.1.0 wasn't pulled for just bug fixes. ;) Too late though, since IBM has already confirmed the existence of the dual-core G5 970MP. :)
[Update 2005-03-16]
Although the above clues have been removed, not surprisingly one can still find several other references to 4 CPUs in the code of the updated MONster.app.
However, with the new 4.1.1 version, all the above clues have been removed. MONster.app now just shows a single CPU for my Cube, with no greyed-out CPU 2, CPU 3, and CPU 4. Interestingly however, the font size in 4.1.1 is the same as in 4.1.0, which means there is sufficient space to display CPUs 2-4, if the cores are present:
It's obvious now that 4.1.0 wasn't pulled for just bug fixes. ;) Too late though, since IBM has already confirmed the existence of the dual-core G5 970MP. :)
[Update 2005-03-16]
Although the above clues have been removed, not surprisingly one can still find several other references to 4 CPUs in the code of the updated MONster.app.
It's official! IBM confirms the 970MP.
IBM has quietly confirmed on its website the existence of the 970MP, by way of a document titled Using Thermal Diodes in the PowerPC970MP® Processor, which begins as follows:
So my friends, no more conjecture. Not only is the 970MP real, it's officially real. ;) Bring on the new Power Macs!
[Update 2005-03-16]
The IBM document has now been pulled. It seems everything is getting pulled right after release these days...
The dual 64-bit core PowerPC970MP™ (970MP) is the next evolutionary step in the PowerPC 970 family of microprocessors. The higher frequency grade versions of the 970MP consume higher amounts of power than earlier IBM microprocessors do, and that can cause temperature issues. Each 970MP processor core contains a thermal diode used to monitor its operating temperature. The thermal diode must be monitored to ensure that the maximum operating temperature of the 970MP is not exceeded.
So my friends, no more conjecture. Not only is the 970MP real, it's officially real. ;) Bring on the new Power Macs!
[Update 2005-03-16]
The IBM document has now been pulled. It seems everything is getting pulled right after release these days...
Monday, March 14, 2005
"Interesting" article about Cell and Apple
Not too long ago, I wrote an article about Cell and Apple, as well as an article about the Xbox 2's CPU and Cell. When I heard tech writer Robert Enderle had written an article called The New Xbox and Apple's Big Decision, I read it with great interest. Unfortunately, it was a dreadful disappointment.
I normally don't write critiques about other people's articles, but this one is simply bizarre. It uses a series of blatantly wrong assumptions to claim that Apple is doomed.
1) The article hints Xbox 2 uses Cell.
2) The article says that since Cell is so different, OSes that run on it have to be so different that any PowerPC support would be through emulation.
3) The article says that IBM is migrating to Cell, and therefore PowerPC is dead.
4) The article says that since PowerPC is dead soon, so is Apple, unless they switch away from PowerPC.
If anything, Microsoft's and Sony's choice (and likely Nintendo's choice) of IBM reaffirms IBM's strong backing of the POWER/PowerPC platform. Indeed, IBM has just recently launched Power.org, a site to promote the architecture.
Quite frankly, even to this undereducated reader, Enderle's article is the worst I've read in a long time about Cell, Xbox 2, and Apple.
[Update 2005-03-14]
It seems the article has been pulled. The link at TechNewsWorld.com no longer works.
I normally don't write critiques about other people's articles, but this one is simply bizarre. It uses a series of blatantly wrong assumptions to claim that Apple is doomed.
1) The article hints Xbox 2 uses Cell.
There is also a processor change that heralds a huge benefit for Microsoft (and likely Sony and Nintendo as well). The change also creates a risk for Apple.That is incorrect. Cell is a chip designed by Sony, Toshiba, and IBM, and Microsoft has nothing to do with it. Actually, Microsoft has not given us the exact specifications yet of the chip for the Xbox 2, but from what we do know it seems its cores are similar to Cell's PPE core. However, what really makes Cell, well... Cell... is everything but PPE.
The Cell Processor
2) The article says that since Cell is so different, OSes that run on it have to be so different that any PowerPC support would be through emulation.
But this doesn't come without a price. This processor is very different from current technology. To make proper use of it, the related OS will need to be largely rewritten, and the only way to run older applications will be under emulation.As I said, Cell's PPE is PowerPC. It uses the same instruction set. Support for Cell's SPEs would be harder, but that's a different kettle of fish. The bottom line though is that there would be no emulation required at all. Cell's PPE is PowerPC, so the OS would just simply have to support one new PowerPC chip, and would maintain full native backwards compatibility with other PowerPC chips.
3) The article says that IBM is migrating to Cell, and therefore PowerPC is dead.
Apple must now realize that the PowerPC will be less interesting to IBM, which is moving its base toward the Cell. Certainly IBM will continue to produce the part for Apple, but it will ramp down development on future versions in order to work on the new Cell instead. At some point, perhaps even within 3 years, IBM will have finished migrating its own platforms to Cell, leaving Apple nearly alone on PowerPC. This is the Apple nightmare scenario.IBM has made no such statement. It could build some some machines with Cell, but could you imagine IBM's big iron POWER servers being replaced by Cell in the foreseeable future? The ironic part is that Cell's PPE unit IS PowerPC.
4) The article says that since PowerPC is dead soon, so is Apple, unless they switch away from PowerPC.
When this point approaches, Apple will have three choices. One is to exit the computer business and concentrate on the more powerful accessories market, focusing on competing with Logitech and Creative Labs rather than with Microsoft. A second option will be to adopt the Cell architecture, but this would signal the obsolescence of the current generation of products and make it more difficult to hold customers. (Remember that the move to OS X Latest News about OS X cost them better than three quarters of the market they had before the migration.) Three is to move to x86, and rumors are once again flying that this choice is being discussed. Such a move, however, would have a dramatic and likely negative initial impact on the installed base.See previous points.
A final option would be to do nothing. This was the strategy that Apple used against the inferior Windows 95 product, and it cost Apple better than 90 percent of the market share it had enjoyed before Windows 95 launched.
If anything, Microsoft's and Sony's choice (and likely Nintendo's choice) of IBM reaffirms IBM's strong backing of the POWER/PowerPC platform. Indeed, IBM has just recently launched Power.org, a site to promote the architecture.
Quite frankly, even to this undereducated reader, Enderle's article is the worst I've read in a long time about Cell, Xbox 2, and Apple.
[Update 2005-03-14]
It seems the article has been pulled. The link at TechNewsWorld.com no longer works.
Saturday, March 12, 2005
Everything Apple's CHUD story on Slashdot, CHUD 4.1 tools pulled
Our Quad Mac & dual-core G5 970MP story from March 10th is now on Slashdot. Everything Apple is honoured. :)
Meanwhile, in addition to a new clue from the CHUD tools we reported yesterday, Macbidouille has added one of their own.
Also, Think Secret reports that Mac OS X 10.4 Tiger will be officially announced on April 1st. This is somewhat surprising since many developers are claiming that Tiger is still a bit rough around the edges. However, Think Secret claims that there is required support of Tiger for a new application to be released at NAB 2005, and that Tiger will ship with new iMacs and eMacs in April as well. (Some readers may note that April 1st is April Fool's Day, but April 1 is actually a very special day for Apple.) This rumoured April Tiger ship date does make one wonder if this could be another hint to imminent new dual-core and quad Mac releases, but Think Secret makes no mention of this.
Lastly, version 4.1.0 of Apple's CHUD tools have been pulled. However, it's not clear if it has been pulled because of all of this dual-core/quad Mac hoopla or if it's for bug fixes. Some online forum posts suggest it may be the latter, with CHUD 4.1.1 to come out soon.
[Update 2005-03-12]
The /. effect:
[Update 2005-03-14]
Thanks to Dave for pointing out that there is indeed confirmation of a major bug in CHUD 4.1.0, necessitating a quick 4.1.1 update. From the Apple PerfOptimization-dev mailing list:
However, the message goes on to say:
It will be interesting to see if the new update still contains references to the 970MP and quad Macs.
Meanwhile, in addition to a new clue from the CHUD tools we reported yesterday, Macbidouille has added one of their own.
Also, Think Secret reports that Mac OS X 10.4 Tiger will be officially announced on April 1st. This is somewhat surprising since many developers are claiming that Tiger is still a bit rough around the edges. However, Think Secret claims that there is required support of Tiger for a new application to be released at NAB 2005, and that Tiger will ship with new iMacs and eMacs in April as well. (Some readers may note that April 1st is April Fool's Day, but April 1 is actually a very special day for Apple.) This rumoured April Tiger ship date does make one wonder if this could be another hint to imminent new dual-core and quad Mac releases, but Think Secret makes no mention of this.
Lastly, version 4.1.0 of Apple's CHUD tools have been pulled. However, it's not clear if it has been pulled because of all of this dual-core/quad Mac hoopla or if it's for bug fixes. Some online forum posts suggest it may be the latter, with CHUD 4.1.1 to come out soon.
[Update 2005-03-12]
The /. effect:
[Update 2005-03-14]
Thanks to Dave for pointing out that there is indeed confirmation of a major bug in CHUD 4.1.0, necessitating a quick 4.1.1 update. From the Apple PerfOptimization-dev mailing list:
To prevent a slew of bug reports and customer pain, let me pre-announce the release of CHUD 4.1.1. :)
We discovered shortly after 4.1 went live that Shark 4.1 can't open saved sessions. Have no fear -- a bug fix is in place, and we'll post a new package as soon as possible.
However, the message goes on to say:
In the meantime, feel free to download the current CHUD 4.1 package and give it a test drive.
It will be interesting to see if the new update still contains references to the 970MP and quad Macs.
Subscribe to:
Posts (Atom)