Slashdot Mirror
Sun & Fujitsu Team On SPARC Chips & System
An anonymous reader writes "Sun and Fujitsu just announced a 20-year partnership to jointly develop SPARC based technology and systems. It looks like the long-predicted partnership that was hinted at earlier has finally come to pass in a much more comprehensive manner than I've heard anyone predict, i.e. not just chips, but a unified range of systems. My guess: Sun drops Ultrasparc III to provide the Throughput computing chips for the low end / web / network stuff, and takes up the Fujitsu provided SPARC64 chips for the high end and workstation market. Will this spark a new RISC renaissance for Sun and Fujitsu? Or is it a last gasp before Opteron / PowerPC / Itanium crush them? I for one will be interested to see what systems and processors come out of this. This could really revitalize the SPARC system market, especially if Sun's work on Throughput computing proves out."
Will Sun be here 20 years from now? Will Fujitsu? If I were a betting man, I'd gamble on the latter more than the former.
This is an interesting deal, and stands to bring much progress in the short term, assuming both parties stick to their commitments. 20 years is a long time, though.
"BSD: Free as in speech. Linux: Free as in beer. Windows 10: Free as in herpes." --Man On Pink Corner in #52607549.
...Japan is called "Land of the rising Sun"
Hehehe... hope they know that Sun wants to give their hardware away.
For thin-client stuff, while low power consumption is a priority, it's not a big enough one to warrant the amount of money that Sun and others have spent on it. Maybe, just maybe, as a spinoff.
These "find a market for our new processor" discussions are getting a little depressing. I remember being excited about the DEC Alpha for embedded applications, but since then it all feels hollow.
1) Sun is having troubles convincing its partners that its multi-core "throughput computing" chip will be competitive. That Gartner report is causing people to ask questions about whether Sun can deliver on its promises. And who wants a 500Mhz 16 core chip anyway? Think of the memory bandwidth problems!
2) Fujitsu Sparc core spanks Suns own core.
My prediction? Sun will abandon its multi-core, asynchronous research pipedreams and farm out all CPU design to Fujitsu. CPU design is a very costly (comoditised) business for Sun to be in, and as Apple have shown its the system that matters, not the processor.
Sun's description of Throughput Computing and their approach of putting multiple processor cores reminds me of what Inmos tried to do with the Transputer before they became STMicroelectronics. The idea was to have many small processors positioned close to each other, communicating between each other closely. I seem to recall seeing transputers on eBay a while back for huge amounts of money. By all accounts, a transputer board was a very useful piece of kit for the right appplications!
> Will this spark
Ha ha ha, very funny.
BoD
Sony just announced they'd stop distributing new Clie outside of Japan...
So, if they left the handheld market to join the Sparc community, is it in order to create servers, or what ?
Trolling using another account since 2005.
My prediction: Sun will return, stronger than you could possibly imagine.
Stick Men
It's more likely Sun will start using Opterons for the low-end. Why? Because (IIRC) Opterons scale much better than Intel chips in a multiple-CPU environment. And that multiple-CPU ability to scale damn near linearly is Sun's real strength in the computer market.
And they want to give that hardware away because they think people are clamoring to pay for the software they put out?!?!!??
Did they have to drive from California to Boston for the ceremony?
;^)
And which one wore the dress?
I have to stop wasting so much time reading Slashdot. It's interfering with my crystal meth addiction.
Because (IIRC) Opterons scale much better than Intel chips in a multiple-CPU environment.
No they don't
Coincidentally, a Fujitsu PRIMEPOWER HPC2500 server ranked #11 on the last Top500 list. That system uses 2,304 SPARC64 V CPUs running at 1.3GHz, delivering an Rmax (tested maximum performance) of 5406 GFlops and an Rpeak (theoretical maximum performance) of 11980 GFlops. The highest all-Sun system on the list? A 672-CPU Fire 15K cluster, way down at #151. Sun does not build seriously big gear. Fujitsu does. That said, systems using competing processor architectures did finish above that Fujitsu. #10 used 1,920 2.4GHz Xeons, #7 used 2,304 2.4GHz Xeons, #6 used 2,816 2GHz Opterons, #5 used 1,936 1.5GHz Itanium 2s, #4 used 2,500 3.06GHz Xeons, and #3 used 2,200 2.0GHz PPC970 "G5" CPUs. (The other systems used Alpha, IBM SP3, or in the case of the Earth Simulator, NEC chips.) So... Fujitsu's big toys aren't going to drop off the bottom of the Top500 list right away, but it looks like other architectures deliver more bang-per-CPU than the 1.3GHz SPARC64 V. (Now, bang-per-CPU-per-GHz, on the other hand, I'm less sure about... I suspect GHz for GHz the SPARC64 V might outrun a Xeon or a PPC970, but might not quite catch an Itanium.) Anyway, Sun's idea about hardware being free should definitely be extended to Fujitsu's supercomputers, yup! :)
They have another market in high end engineering desktops. For people who design chips and other detailed components and need to simulate them there is still a market for their work stations.
Now, the other chips are catching quick on this so they need to stay ahead or they could loose that market too.
Evolution or ID?
The Business Software Alliance believes new technologies should enhance ways to access and distribute copyrighted works legally, not illegally.
Couldn't agree more, old chap! Three cheers for the BSA!
Baloney. I've had power drop out and kill Sun machines and they just come right back. Are you talking about file system check? Just turn on logging in /etc/vfstab and even that goes away, just like the same reason you use ext3 instead of ext2 on Linux.
when yout go the multi-threaded, CMT, etc.. route then Amdahl's law trumps Moore's law.
No, I was talking about hardware failures because of power problems. Where I went to college and where I wonk now we have a problem with hardware breaking from power failures.
Evolution or ID?
Stop it, you're punning me to death. But really though, will the benchmarks from the new systems be fiery or all wet? How heavy are the servers, are they any lighter? And will the chips light your boxes afire or will they be different from the flaming Xeon? Will it be a match for Opteron?
Even those who arrange and design shrubberies are under considerable economic stress at this period in history.
And how fast is a "shared memory" cluster when processor 34 updates a value that is cached by processor 91? Will processor 91 even know about the change to the value it's holding? This "cache coherency" issue is a huge problem in multiprocessor architecture, and Sun has solved it - and solved in so that it scales damn near linearly up to a few hundred processors at least.
That is a slick move- offer free hardware- and then team up with a hardware company to pay for it. Brilliant.
That's true of a lot of computers, not just Sun. that's why they invented these amazing things called UPS'
Besides, power failures shouldn't happen; you should have UPS on all important servers so power failures shouldn't be a problem at all.
So yesterday there was the post about hardware being free and today there's news about Sun partnership for UltraSparc. Make up your freaking mind SUN. I still like SUN hardware for dense deployments where you got tons of racks, since Intel and AMD both run considerably hotter than UltraSparc. Sun needs to make up its mind about whether they believe UltraSparc has a future and stick to it. PC hardware is still isn't as reliable as high end Unix, so it's stupid to drop their R&D for high end systems. Some things simply scale better vertically than horizontally.
I wonder if funding a kind of open-sourced game development platform wouldn't help the hardware manufacturers (e.g. Sun in this case for the processors and ATI etc. for the graphics) sell their wares. All 3D-engine stuff wide open and free, so that a prospective game maker would not have to buy an engine license.
Problems of course:
- need an installed base to sell enough games
- state of the art engine does not grow on trees
- willingness of hardware types to work together
Possible Pros:
- open standard encourages performance improvements for the next generation that don't break older applications
- massively distributed bug fixing
today that have a mandatory need for
"some sophisticated algorithms for migrating threads to the most appropriate processor based on things like memory locality and load"
Get a clue. Ever hear of "web servers"? OLTP? Real-time video processing?
I really don't think Oracle doesYou went two words too far.
Microsoft's
software doesn't (doesn't even run on sparc);Microsoft's offerings are a joke in the world of high-performance computing. The fact that you even bring Microsoft into a discussion on HPC immediately puts you into the "clueless but still hyping my MCSE" corner.
I don't think OpenOffice needs it
Do you really think OpenOffice of all things defines what high-performance hardware needs to do? Earth is defintely calling, but I don't think the signal is being received....
To use a simile, if your analysis is correct,
sun's os+hardware is like buying a $300,000
Feraris for a 5 mile daily commute.
Only an idiot with more money than common
sense would spend that much money when
they could spend $10,000 on a cheap and reliable
commuter car for the 5 mile daily commute
and spend the rest of the money on a house/vacation/girlfriends/etc.
What kind of moron thinks he can show up at the Indy 500 in a fricken' Yugo?
You must think the desktop is where real computers are. Put you fricken' toy aside.
With the Millenium project dead and buried Sun is relying on Ultra SPARC III to carry the comapny until "Mid 2006" when Fujitsu's and Sun's products will be shared between them. The current Niagra design (from Afara) is so low end that Sun will be luck to sell a few hundred systems based on it. If they didn't need a throughput computing product on the market RIGHT NOW they would have killed Niagra already.
When the inevitable schedule slips on Niagra II and Rock come to light (the original Niagra from Afara was "almost done" when Sun bought them two years ago, it's only just taped out) Sun will have no choice but to fall on it's sword and admit defeat. The company might survive if it can convince enough customers to recompile and move to Opteron based systems while sticking with Solaris, but that's going to be a hard sell when they can recompile for linux and not be locked into Sun's software/services stack.
It is, if you look under the hood:
bash-2.05$ uname -a
SunOS XXXXXX 5.9 Generic_112233-10 sun4u sparc SUNW,Ultra-Enterprise
bash-2.05$
Say maybe like Solaris Zones due for release on Solaris 10 later this year?
From the article:
I have been playing around with Zones on Solaris Express (Solaris 10 pre-release) and they deliver on their performance promise while easliy isolating applications.
IMHO this is a wise move by Sun. They can not only share their (considerable) R&D costs with Fujitsu, but they can take advantage of some of the nice performance gains FS have been able to make with SPARC64 (FS machines are big in HPC environments, where Sun doesn't have much traction). This will help their margins and maybe finally they can be profitable again. The other thing is that it gives the SPARC platform a united front and potentially greater market coverage (especially in Europe, where FS are very strong). Maybe there is life in the old dog yet. I hope so, the server market needs companies like Sun. Leaving it to Intel and Microsoft would be a backward step for the industry.
I don't think the Itanium is going to be crushing anything in it's short lifetime. However, the PowerPC/Opteron chips are putting the smack down quite nicely about now. We need them to bring back that Open spec for PPC hardware so we can get some serious speed and off of Intel..... !
-m
http://www.invisik.com
Sun and Fujitsu just announced a 20-year partnership to jointly develop SPARC based technology and systems.
They have had a partnership for 20 years - they aren't announcing a new one.
My guess: Sun drops Ultrasparc III
Sun is already shipping the Ultrasparc IV. Nice guess!
to provide the Throughput computing chips for the low end / web / network stuff,
They have already announced that this is exactly what they are going to do. Again: nice guess!
and takes up the Fujitsu provided SPARC64 chips for the high end and workstation market.
Yesterday's announcement was all about using SPARC64 on the high end. Usually the trick is reading between the lines - not reading the lines themselves.
Sun also announced that they will be using Opterons in their new workstation line - not SPARC64.
Will this spark a new RISC renaissance for Sun and Fujitsu? Or is it a last gasp before Opteron / PowerPC / Itanium crush them?
Itanium has gone white dwarf. The only thing it will be crushing is itself.
Opteron is not going to crush Sun. They have announced that they are shipping multiple Opteron boxes (1-8 way servers and 1-2 way workstations).
This could really revitalize the SPARC system market, especially if Sun's work on Throughput computing proves out.
This doesn't even make sense. The Fujitsu/Sun machines are not the Throughput Computing systems that Sun has been talking about for months. Throughput compututing is Niagara/Rock - the Sun-only CPUs.
- Old Man of the Mountain ---- "I want to disturb my neighbor"
To put the matter simply, what killed Sun Microsystems is a pathetic engineering team in the microprocessor division. With the exception of the UltraSPARC I and II, all the other processors were poorly designed and managed. What is unique about Sun's microprocessor division is that the managers consistently and actively hired H-1B workers from Taiwan and India. Foreign engineers were the rule, not the exception.
Not so with Fujitsu. Admittedly Fujitsu had a similar problem with the SPARC64-I, SPARC64-II, SPARC64-III, and SPARC64-IV because all these processors were developed at HAL in Campbell, California. HAL also hired mainly foreign engineers, and previous generations of SPARC64 sucked. Then, Fujitsu became tired of this nonsense, shutdown HAL, and fired everyone. Fujitsu then developed the SPARC64-V entirely in-house, using only Japanese engineers. No foreign engineers.
There is a myth that, somehow, tech companies absolutely need H-1B workers. Well, now we have yet another example of why that myth is just a myth. SPARC64-V built by native engineers crush UltraSPARC III built by H-1B engineers.
To understand how pathetic Sun's microprocessor engineers are, we in the server division actually had servers ready to accept the new UltraSPARC III by the end of 1999. Unfortunately, the processor team was two years late. So, our test machines sat idle.
Note that the server division is not dominated by H-1B engineers. The server division and the microprocessor division are two different worlds: first world versus 3rd world.
I, for one, am glad that we are relying on Fujistu. Its processors are much better designed and built than Sun's own processors. I am glad that Fujitsu will soon OEM high-end servers to Sun. Sun will stop designing and building high-end servers in 2006. (I work on the low-end servers running x86.)
I simply do not see Niagara and Rock as the savior of the company because those designs are well-known public knowledge. Check out Professor Kunle's Hydra work: it is 70% of Niagara. Intel has now embraced the Hydra work and will produce an x86 chip based on Hydra.
Here's a dumb question: Which company will build the fastest, highest performance multi-core chip based on Hydra? Intel or Sun?
Ohhh, there is not better than Tomasulo or Scoreboard's Algorithm out-of-order execution.
Why not a VLIW 64-bit architecture with 2000-pins chip and a good VLIW64 64-bit compiler?
where is the LaMerced-Killer?
open4free ©
...welcome our new Sun&Fujitsu overlords!
The real upside of of zones is when combined with N1 Grid. I know it sounds really amorphous, but with zones it really starts to make sense. Imagine rolling out a new zone or dozens when needed across a datacenter full of stock Solaris 10 machines (x86 or SPARC). The storage is on EMC or Hitachi, so you just hand out LUNS like candy for the zone's "/" with predefined pkgs and patches and you have a really dynamic environment for Oracle 10g, webservices, etc.
This is the story that needs to be told, but I sometimes think Sun is using Novell's old marketing team.
DISCLAIMER: I work for Sun but I try not to drink the company Kool Aid.
_damnit_
It's my job to freeze you. -- Logan's Run
I don't get it. Low end machines are so small that scaling isn't an issue. Sun will use Opteron in the low end simply because it is cheap.
hardware breaking from power failures
You're not alone. While I feel this is uncharacteristic of Sun hardware in general, I saw the same thing happen to Sun hardware at one company. They switched to Fujitsu hardware (which implements the SPARC IV platform and is Solaris compatible), and have had no similar problems running on the faster Fujitsu machines.
Fijutsun anybody?