Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Sneak Peek At Sun's SPARC Server Roadmap

Soulskill posted more than 5 years ago | from the measuring-stick-for-oracle dept.

Sun Microsystems 113

The folks at The Register have gotten their hands on Sun's confidential roadmap from June, which outlines the company's plans for SPARC product lines. The chart has some basic technical details for the UltraSPARC T-series and the SPARC64 line. The long-anticipated "Rock" line is not mentioned. "We can expect a goosed SPARC64-VII+ chip any day now, which will run at 2.88 GHz and which will be a four-core, eight-threaded chip like its 'Jupiter' predecessor. This Jupiter+ chip is implemented in the same 65 nanometer process as the Jupiter chip was, and it is made by Fujitsu, a company that is in the process of outsourcing its chip manufacturing to Taiwan Semiconductor Manufacturing Corp. ... not only has Sun cut back on the threads with [the 2010 UltraSPARC model, codenamed Rainbow Falls], it has also cut back on the socket count, keeping it at the same four sockets used by the T5440 server. And instead of hitting something close to 2 GHz as it should be able to do as it shifts from a 65 nanometer to a 45 nanometer process in the middle of 2010, Sun is only telling customers that it can boost clock speeds to 1.67 GHz with Rainbow Falls."

cancel ×

113 comments

Sorry! There are no comments related to the filter you selected.

Overcome by events (2, Insightful)

russotto (537200) | more than 5 years ago | (#29394905)

The only things on Sun's roadmap now are signs to the effect of "Road Closed 1000 feet".

Re:Overcome by events (1)

TheRaven64 (641858) | more than 5 years ago | (#29395137)

I guess that's why Oracle are running ads at the moment saying that they plan on spending more on SPARC and Solaris development than Sun ever did...

Re:Overcome by events (2, Insightful)

timeOday (582209) | more than 5 years ago | (#29395207)

But why? 10 years ago I thought sharing an 8 CPU Sun with a big devel team was a privilege. Now any decent Dell workstation has that. What does SPARC have over Intel? (No vague claims of superior "throughput", please!)

it has throughput and power efficiency (5, Informative)

Anonymous Coward | more than 5 years ago | (#29395725)

But why? 10 years ago I thought sharing an 8 CPU Sun with a big devel team was a privilege. Now any decent Dell workstation has that. What does SPARC have over Intel? (No vague claims of superior "throughput", please!)

It has throughput. Back in 2006, when the first T2000 was released, a dual Xeon could handle 980 req/s from Apache and the T2000 could handle 15,000 req/s:

                http://www.stdlib.net/~colmmacc/2006/03/23/niagara-vs-ftpheanetie-showdown/
                http://www.stdlib.net/~colmmacc/2006/03/27/niagara-benchmarks-update/

At the same time the Xeon used a peak of 2.2 Amps, while the T2000 peaked at 1.2 A. Things have only gotten faster.

Throw-in on-board crypto, and you can do AES-128 at 38.9 Gb/s with a single socket (eight core) T5220:

                http://blogs.sun.com/bmseer/entry/ultra_fast_cryptography_on_the

A T5440 can do 22,932 MB/s (183,456 Mb/s = 179 Gb/s):

                http://blogs.sun.com/yenduri/entry/t5440_crypto_performance_numbers

If you're a site that cares about SSL/TLS, how many x86 machines would need to buy, maintain, and cool to handle that load? How many F5 load balancers/SSL accelarators would you purchase? According to F5's own data sheet, the 8900 (with dual 850W P/S) can handle 9.6 Gb/s--and you still have to buy web servers on top of that (more power)

So the T5120 can do roughly four times the raw encryption rate, uses dual 720 W P/S, and also do work as web servers. You're also using less rack space.

Let's also compare to AMD-based systems (which Sun also sells):

                http://blogs.sun.com/bmseer/entry/web2_0_consolidation_sun_sparc

Now the Niagara (UltraSPARC-Tx) CPU isn't good for every work load out there, but if it's highly parallel then it's something that you should be looking.

Re:it has throughput and power efficiency (1)

JAlexoi (1085785) | more than 5 years ago | (#29395941)

However any floating point operations make the T2000 look like Pentium 2.

Re:it has throughput and power efficiency (1)

Tubal-Cain (1289912) | more than 5 years ago | (#29396011)

So don't buy one if you need floating-point operations.

Re:it has throughput and power efficiency (2, Informative)

Anonymous Coward | more than 5 years ago | (#29396793)

That was only true for the first chip. The T2 series has 1 FPU per core.

floating point (0)

Anonymous Coward | more than 5 years ago | (#29397315)

However any floating point operations make the T2000 look like Pentium 2.

Apache, lighthttp, nginx, Varnish, Squid, and memcached generally don't use much floating point AFAIK.

Also, the T2000 is set to be retired RSN, and so the smallest box available is the T5120 which has a lot more FP units.

Re:it has throughput and power efficiency (1)

NotBornYesterday (1093817) | more than 5 years ago | (#29401519)

No. The later generations of Niagara procs from the last 2 years or so have 1 FPU per core and perform well under floating point loads. There are still some original T2000 systems out there, but they are specifically for business-oriented apps with mostly integer operations.

Oracle + Niagara = expensive? (2, Interesting)

TheLink (130905) | more than 5 years ago | (#29396507)

> Now the Niagara (UltraSPARC-Tx) CPU isn't good for every work load out there, but if it's highly parallel then it's something that you should be looking.

If Oracle still charges per core, the Niagara approach of many core CPUs could be more expensive.

Looking at the roadmap they seem to be going fewer cores, or at least sticking with 8.

As for power consumption, I wouldn't bet on the Intel x86 always consuming more power than a SPARC for the same performance. They are a scary competitor. They keep introducing consumer grade x86 cpus that are more powerful and yet consume less power.

Can Sun/SPARC keep ahead of them? They might only be ahead in SSL/TLS. And if that becomes a big enough demand, some taiwanese/chinese company start producing cheap pcie cards to do that :). Or Intel could decide to use some transistors to do it - they have lots of transistors to play with on their chips, it's just a matter of priorities.

Re:Oracle + Niagara = expensive? (2, Informative)

joib (70841) | more than 5 years ago | (#29396647)


Can Sun/SPARC keep ahead of them? They might only be ahead in SSL/TLS. And if that becomes a big enough demand, some taiwanese/chinese company start producing cheap pcie cards to do that

Crypto accelerator cards have been available for a long time. Don't know about the price though.


Or Intel could decide to use some transistors to do it - they have lots of transistors to play with on their chips, it's just a matter of priorities.

See "Sandy bridge", Intel's next 32nm chip, due Q1 2011, will have extra instructions for AES.

Re:Oracle + Niagara = expensive? (0)

Anonymous Coward | more than 5 years ago | (#29397423)

Crypto accelerator cards have been available for a long time. Don't know about the price though.

Yes, Sun sells them. The on-CPU crypto trounces them in performance because you don't have to go through the bus.

See "Sandy bridge", Intel's next 32nm chip, due Q1 2011, will have extra instructions for AES.

Great. Nothing I can use now.

Also note that each Niagara 2 CPU has two 10 GigE interfaces right on each die, so if you need to push out to the network (especially encrypted), you can go from encryption unit to networking unit without touching the memory controller/DMA. So your bus bandwidth is concentrating on feeding the exec units from memory and not having to shuffling I/O back and forth.

I have yet to see any numbers that indicate that UltraSPARC-Tx can be beaten by the same number of Intel- or AMD-based boxes when it comes to highly parallel workloads. (The main provisos are with regards to floating point. :)

Personally I also prefer the console LOMs of SPARC machines more than the BIOS-redirect of most of the x86 hardware. IPMI and DRAC/iLO is fine for a lot of things, but going to serial (or SSHing in and getting the console redirect) is a very handy all-else-fails mechanism.

Re:Oracle + Niagara = expensive? (1)

ToasterMonkey (467067) | more than 5 years ago | (#29403319)

I have yet to see any numbers that indicate that UltraSPARC-Tx can be beaten by the same number of Intel- or AMD-based boxes when it comes to highly parallel workloads. (The main provisos are with regards to floating point. :)

At spec.org you can see the previous generation of Xeons caught up and started peeling away from T2 performance, ad the new Nehalem Xeons absolutely throttle everything else out there (similar to how Sun's CMTs leapt ahead at launch). The most interesting thing I learned from staring at spec.org charts is that sooner or later a vendor will pull away from the competition like this, and it never lasts. Sort the charts by available date, and you can see Sun/Intel/AMD pulling away at times. I think Sun and AMD will catch up, but it might be a year out. Took Intel & AMD that long to catch up to the first CMT's anyway.

Personally I also prefer the console LOMs of SPARC machines more than the BIOS-redirect of most of the x86 hardware. IPMI and DRAC/iLO is fine for a lot of things, but going to serial (or SSHing in and getting the console redirect) is a very handy all-else-fails mechanism.

SUN's ILOMs for x86 and more recently T5xxx+ SPARC servers do bios redirection over serial (x86). On either system, ssh ilomaddress, start /SP/console
It feels a little clunky on x86, but does what you expect on SPARC. I think it's slowly replacing their ALOMS. The Java based remote console GUI (x86 ILOM only) beats the pants off DRAC's ActiveX crap. That's got to be a little painful for those running Linux on Dells :P

Re:Oracle + Niagara = expensive? (0)

Anonymous Coward | more than 5 years ago | (#29398431)

Exactly Oracle needs to make the Niagara more competitive for their software. We are a Sun shop and it's cheaper for us to buy an M5000 + Oracle than a T5220/40 + Oracle. I would love to buy Niagara for our middle tier, it's just that Oracle would skull us over the cost. They need to move back to the old T2000 model .25 per core or now .10 per core for Niagara and bump the core charge on Power Series...if they are serious about the long term future of Sparc. Larry should do this today just like the WSJ ad should have been put out months ago

Re:Oracle + Niagara = expensive? (1)

Zero__Kelvin (151819) | more than 5 years ago | (#29399365)

"If Oracle still charges per core, the Niagara approach of many core CPUs could be more expensive."

Just run it in a single core Virtual Machine, e.g. VirtualBox [virtualbox.org] .

Re:it has throughput and power efficiency (1)

g00ey (1494205) | more than 5 years ago | (#29397637)

What about Intel's Itanium2 chips?

so why doesn't google use sun? (0)

Anonymous Coward | more than 5 years ago | (#29397879)

if i take your argument at face value, it would be a no-brainer.
clearly you're presenting half the facts. also a comparison
of 2009 hardware would make more sense.

Re:it has throughput and power efficiency (0)

Anonymous Coward | more than 5 years ago | (#29399329)

I'm not about to use server cpu for SSL/TLS.

Re:it has throughput and power efficiency (1)

k8to (9046) | more than 5 years ago | (#29402361)

Now the Niagara (UltraSPARC-Tx) CPU isn't good for every work load out there, but if it's highly parallel then it's something that you should be looking.

Highly parallel with *low* cpu needs.

Niagara is good at dispatch and switching, but not computation.

Re:Overcome by events (1)

multiplexo (27356) | more than 5 years ago | (#29403341)

Because it has incredibly superior throughput. I know that there are lots of fanbois out there who are hung up on substitutes for dick size such as clock speed or number of cores, but the throughput on a cheap Sun T2000 kicks the living shit out of anything Dell has in a comparable size and price. If you're running anything that needs to push a lot of bits back and forth throughput is important. Go ahead and try running a NetBackup master server on an Intel box running Linux. It can be done, but the performance will suck ass compared to a Sun or HP box.

Any moron can put together systems that have lots and lots of CPUS and lots and lots of cores and lots and lots of Mhz and which deliver impressive SpecInt and SpecFPU numbers and many have. However engineering a system which has great throughput is a lot harder to do.

I'm cautiously enthusiastic about Oracle's takeover of Sun and hope that they don't fuck things up. Not having to deal with Oracle's idiotic processor multiplier licensing price for Sparc CPUs will be a huge improvement.

Re:Overcome by events (1)

rubycodez (864176) | more than 5 years ago | (#29401397)

no, oracle said they would spend more "than sun does now". which is next to nothing for R&D since their sales have tanked.

Re:Overcome by events (3, Insightful)

onionman (975962) | more than 5 years ago | (#29395147)

Unfortunately, that's looking more true every day. I remember running a network of Sparcs and bragging to my family members about how they (the Sparcs) were sooo much more powerful than PCs that we had in our homes. Seven years later I was replacing all our Sparcs with x86_64 Linux boxes... too bad Sun just couldn't keep up with hardware development. It would be nice if Oracle really did ramp up hardware R&D for Sun, but I can't see those announcements being anything more than reassurances to nervous enterprise customers.

No Economies of Scale (2, Interesting)

reporter (666905) | more than 5 years ago | (#29395245)

In order for a microprocessor to be financially successfully, it must enjoy large economies of scale. That Intel can sell essentially the same design (of the x86) in multiple forms to hundreds of millions of customers means that Intel can afford the massive research and development that is necessary to design the typical x86 chip.

By contrast, though Sun Microsystems often boasted that it has -- actually, had -- the largest microprocessor team after Intel, the team could not design a chip that sold to hundreds of millions of customers. They numbered only several hundred thousand. That sell rate could not pay the salaries of the 2nd largest microprocessor team in the world.

Here is another example of Sun arrogance. In the 1990s, Sun could have easily built their company on the unglamorous ARM RISC processor, but Sun management wanted to exhibit the "pride" (and arrogance) of homegrown technology. So, the management spent billions of dollars on designing one of the worst microprocessors in the industry. The UltraSPARC III was overbudget, was late (to market), and underperformed its peers.

The Sun has set. Good riddance to arrogance.

P. S.

Yes. The UltraSPARC I and II were okay. However, they were not stellar. What helped their sales was mating them to the server box, E10000, that Sun luckily acquired from SGI/Cray. However, starting from the UltraSPARC III (an utter fiasco), the processors were so horribly underperforming that even an outstanding server design could not compensate for the mediocre performance of its processor.

Re:No Economies of Scale (1)

hemp (36945) | more than 5 years ago | (#29395319)

Believe it or not, but back in 1990, I worked on a Sun workstation with an Intel 80386 processor that ran SunOS and *DOS* ( http://en.wikipedia.org/wiki/Sun386i/ [wikipedia.org] .

Re:No Economies of Scale (1)

that this is not und (1026860) | more than 5 years ago | (#29395339)

I have a Sun 386i too.

However, at the time, Sun didn't have the Sparc yet, and were looking for the migration path forward from the 68000.

Now, I can run DOS on my Macintosh SE/30 (bochs on NetBSD) and have.

Re:No Economies of Scale (4, Interesting)

Jeremy Allison - Sam (8157) | more than 5 years ago | (#29395375)

Yeah, lots of the original Samba code was written on Sun 386i's. Ah, memories :-).

Jeremy.

Re:No Economies of Scale (1)

TheRaven64 (641858) | more than 5 years ago | (#29397167)

In the 1990s, Sun could have easily built their company on the unglamorous ARM RISC processor, but Sun management wanted to exhibit the "pride" (and arrogance) of homegrown technology

And they were right to do so. ARM focussed entirely on the embedded market and left companies like Acorn in the cold for workstation chips a few years later. They were lower power, and maybe cheaper too, but they were much slower than anything else on the market; much slower than. Sun's mistake was to choose not to compete with ARM; they had low-power SPARCv7 designs, but never pushed them into the mass market. If they'd sold a stack based on the *7 prototypes (low power SPARC+Solaris running in 1MB of RAM with execute-in-place) then they'd be a key player in a growing market, not a tiny-and-shrinking one.

Re:No Economies of Scale (1)

drinkypoo (153816) | more than 5 years ago | (#29397443)

Sun kicked super-lower-power microsparcs out into the market; the market Did Not Want them. They cost too much (surprise!) and weren't as fast as offerings coming from basically everyone else at the time.

Re:No Economies of Scale (0)

Anonymous Coward | more than 5 years ago | (#29401093)

You know, the x86 chips are general purpose processors, that are intended to be run on general purpose machines, this results in selling hundreds of m,illions of units because the marker for GP hardware is huge.

You're operating under the impression that UltraSparc is also designed for general purpose computing, and s such should also be selling hundred of millions of units - but it isn't. It's designed for specific purposes, used for those purposes, and in a much higher margin market. Consider than even Intel doesn't position the x86 as a competitor to Sparc in Sparc's home market - they have Itanium for that, and I'd hate to break it to you, but not only does IA-64 not sell in the hundreds of millions, either, it is outsold by the Sparc T2 by a considerable margin.

HP's PA-RISC and IBM's Power don't sell in high volume either, so I really don't get where you're going with this. And Zilog z80 sales utterly dwarf x86 sales, not all markets are the same, you know, and really, volume alone says very little when you don't take margins, nor the size of two separate markets into account.

Re:No Economies of Scale (1)

segedunum (883035) | more than 5 years ago | (#29401833)

You're operating under the impression that UltraSparc is also designed for general purpose computing, and s such should also be selling hundred of millions of units - but it isn't.

The problem is that's what UltraSPARC ended up competing against, despite protestations to the contrary from a lot of people. SPARC machines have been replaced by the boatload with x86 and x86_64, which is why the architecture is in trouble. When you benchmark an UltraSPARC III as I did with pystones against a 1.4GHz Athlon and find the Athlon is between two and three times as fast then you're entitled to ask what you're spending the money on.

Re:No Economies of Scale (0)

Anonymous Coward | more than 5 years ago | (#29402655)

You're just plain wrong, and it shows in the way . Volume is everything today. Volume decides what lives and what dies.

The days of the fat profit margins are gone; Sun was just too stubborn and resisted going mass-market.

SPARC systems were not made dirt-cheap, as cheap or cheaper than i86pc hardware, and that severely pissed off former sparc customers like myself.

Re:Overcome by events (1)

Taco Cowboy (5327) | more than 5 years ago | (#29396729)

For MIPS and Digital, they have hit the end of the road

For SUN, their end of the road is near

And I am afraid the same would be for the now fabless AMD

One day in not that long in the future we gonna wake up to the fact that only IBM, Intel and some Taiwanese companies (Nvidia, VIA, TSMC) gonna be the only one left still making power processors for the world

And if I am not wrong, IBM may end up not making chips as well

Re:Overcome by events (1)

thogard (43403) | more than 5 years ago | (#29397299)

I want a SPARC IIIi running at 65 or 45nm with modern gig ethernet. It would be far more than I need for my apps. I would like it in a box the size of the Netra X1 and running any OS the old ones could run like Solaris 9. It would be cool if they were $1000 each like the old X1 or V100. I might buy several hundred in that case. Meanwhile I'm buying old X1 systems and putting in SSD and replacing fans and power supplies and hoping for the best.

And I have loads that are faster on the old X1 than the t1000 so I don't want that junk.

Not a good idea to publish this (3, Informative)

harmonise (1484057) | more than 5 years ago | (#29394931)

The folks at The Register have gotten their hands on Sun's confidential roadmap from June

If it's confidential then the Reg shouldn't publish the details. Unless they want to give Sun's competitors a leg-up. I'm sure Sun's competitors marketing teams are happy to have this. [sigh]

Re:Not a good idea to publish this (1)

Henriok (6762) | more than 5 years ago | (#29395015)

In this business, having a two months head start is..nothing. I bet HP's and IBM's marketing teams are launching whatever is in the pipe according to nuances Sun's roadmap. It must be the reason Tukwila is slipping year after year after year.. or not.

Re:Not a good idea to publish this (3, Interesting)

icebike (68054) | more than 5 years ago | (#29395043)

Then Sun should, in fact, keep it confidential.

I'm betting it was leaked to give some assurance to the customer base that there will actually BE a Sun in the future.

Re:Not a good idea to publish this (3, Insightful)

eln (21727) | more than 5 years ago | (#29395267)

It's not necessarily Sun that leaked it. Hardware manufacturers (and software houses, for that matter) routinely show large customers their roadmaps under NDA. It's entirely possible some less than scrupulous employee of one of their big customers leaked it, in violation of their NDA.

Re:Not a good idea to publish this (1)

mehemiah (971799) | more than 5 years ago | (#29395703)

I think this was a leak like some campaign and whitehouse leaks are. It's possible that they unofficially released this on purpose to gauge the public response. This, as aposed to being an Apple leak which is simply a ploy to plug leaks. Then finally there are leaks which, like predictions and interpretations of Lost, are simply guesses that they got right because when enough people are guessing about the same thing, someone statistically has to get it right. (like the iPhone) I still call BS on the people STILL saying there will be an Apple netbook, THERE ALREADY WAS ONE, it was the Macbook Air, GET OVER IT! ok, got a little off topic there sorry, venting.

Re:Not a good idea to publish this (3, Insightful)

hairyfeet (841228) | more than 5 years ago | (#29396029)

Which anybody who actually thought about it for more than 2 seconds would know that Oracle would be keeping SPARC and Solaris around for a LOONG time. I mean lets be honest here: Like Bill and Steve old Larry may be a bit of a bastard, but he is a bastard that knows how to get his ROI. As other posters have pointed out Solaris + SPARC equals high throughput in specialized tasks (like say...an Oracle database) and more importantly to Larry he now controls the whole smash, from the OS down to the hardware.

With Linux it wasn't like he could call up Linus and demand he rewrite the kernel to give Oracle maximum throughput, but with Solaris and SPARC he can have the direction of the entire thing shaped by HIS design, and towards making it the fastest platform for Oracle possible. And of course by owning the whole thing it will make many an admin and PHB happy, as there is only one company to call if things go wrong and none of this "it is the other guy's stuff!" blame passing.

So I doubt VERY seriously you'll be seeing anything like killing Solaris or SPARC. More likely Larry will make damned sure that future development will be tailored to Oracle, making Solaris+SPARC+Oracle the preferred platform for anyone running Oracle. And thus making Larry a whole hell of a lot more money in the process. It just makes good business sense.

Re:Not a good idea to publish this (1)

drinkypoo (153816) | more than 5 years ago | (#29397417)

Which anybody who actually thought about it for more than 2 seconds would know that Oracle would be keeping SPARC and Solaris around for a LOONG time.

I don't agree. They might TRY to, but it looks very much like SPARC is out of steam. POWER is beating it like a pinata.

Sun couldn't keep SPARC on top no matter how they tried. Oracle pledges to spend more money, but it's not clear where the money will come from or that it will do any good.

Re:Not a good idea to publish this (0)

Anonymous Coward | more than 5 years ago | (#29398721)

I don't agree. They might TRY to, but it looks very much like SPARC is out of steam. POWER is beating it like a pinata.

There are people who would disagree with you (with benchmarks):

http://blogs.sun.com/bmseer/tags/ibm

Re:Not a good idea to publish this (1)

raftpeople (844215) | more than 5 years ago | (#29399257)

Here is a problem with T2: it can't scale.

The largest box is 4 proc. Power procs are designed to scale and they sell 32 and 64 proc systems (as does HP with Itanium). T2 doesn't go here and Sparc (as in UltraSparcs) gets beat easily.

Re:Not a good idea to publish this (1)

asaul (98023) | more than 5 years ago | (#29404163)

What workload needs 64 POWER or SPARC procs anymore? More often than not if it needs that much CPU it is horizontally scalable anyway, in which case buying 2x as many T series boxes would be cheaper anyway. Most of the time the reason you have boxes with 64 CPUs installed is for partitioning with LPARs or domains.

And for scaling it all depends what you are doing with the box. We have an application which consumed a full 48 core E6900 (i.e the box was 100% on CPU) because it ran all its components on the one box. We moved a component that consumed 50% of the load onto a single T5240. The T5240 was only consuming 15% CPU with improved response time (granted it was a Borland java application which suited a T-series box a lot better).

For the cost of a E6900 uniboard, we could buy 2-3 T5240s to replace the E6900 and the T5240s would handle 6 times the load.

Re:Not a good idea to publish this (0)

Anonymous Coward | more than 5 years ago | (#29401173)

and towards making it the fastest platform for Oracle possible

And Larry's got the advantage that it ALREADY is the best platform to run Oracle on, in terms of performance, throughput, vertical scalability and even price/performance.

Also, because it's bound to come up, because people can't seem to distinguish between horizonal scalability (scaling out - more sockets, clustering. etc) and vertical scalability (scaling UP - more throughput on fewer sockets) - They're bound to mention that Power and IA-64 machines come with more CPUs per board (than the T2 machines, they don't like to mention Sun/Fujitsu's M-class behemoths) neither Power nor IA-64 comes close to Sparc's raw throughput per socket.

I mean, the fact that Sparc64 VIII/fx dishes out 126 GFLOPs or that the Sparc64 VII based M9000(64) currently holds the record (at over a teraFLOP) for the most powerful single-board computer system is nothing to take lightly.

Re:Not a good idea to publish this (0)

Anonymous Coward | more than 5 years ago | (#29403097)

what good is it when it ends up draining away the profits he makes from selling to us, the other platform oracle folks?

It's not easy being a hardware company, especially now when everybody wants everything at a wallmart special price.

This is hardly a secret (4, Informative)

Freaky Spook (811861) | more than 5 years ago | (#29395103)

Sun have been providing theses details to their Partners at the Sun Partner Advantage Summits, I got this info last month.

Plus Sun Partners just have to contact their Sun Sales managers and just ask for a Roadmap Session(Under Signed NDA)

The Register are just publishing what already is pretty common knowledge amongst most people working with Sun/SPARC hardware already, it won't give their competitors a huge advantage at all, the fact that Sun are already revealing this stuff to their wide partner network means that the development of it is well and truly in its final stages, and if their competitors are finding this out through The Register, then they really are not doing their jobs properly.

Re:This is hardly a secret (1)

NoYob (1630681) | more than 5 years ago | (#29395281)

...if their competitors are finding this out through The Register, then they really are not doing their jobs properly.

I had this image of Intel, Motorola, AMD, (I can't think of more) sending really hot Russian women to go and seduce the SUN engineers getting them to divulge everything. After I stopped my 007ish fantasy, I realized all they'd have to do is send in a pretty lady and have her just say "Hi" and those engineers would divulge everything - they are geeks after all.

Re:This is hardly a secret (0)

Anonymous Coward | more than 5 years ago | (#29395397)

And where would the guys at The Register get a pretty lady?

Re:This is hardly a secret (0)

Anonymous Coward | more than 5 years ago | (#29395497)

At the pub?

Re:This is hardly a secret (1, Funny)

Anonymous Coward | more than 5 years ago | (#29395603)

Idiot. Sun has a lot of female geeks, too.

Re:Not a good idea to publish this (0)

Anonymous Coward | more than 5 years ago | (#29395151)

Who says it's real?

It could very well be published by IBM, Microsoft or one of a dozen other companies for shits and giggles...

Peak??? (5, Informative)

hguorbray (967940) | more than 5 years ago | (#29395031)

I know this is a dark age for literacy, but s/b peek -ya know? like PEEK and POKE???

I'm just sayin'

Re:Peak??? (1)

Cambo67 (932815) | more than 5 years ago | (#29395051)

They may be referring to an unexpected hill that the cartographers missed.... :)

Re:Peak??? (1)

TheLink (130905) | more than 5 years ago | (#29396431)

Maybe they meant it's a sneaky little peak before the long steep downhill in the roadmap.

"Peak" SPARC just like "peak oil".

And costs are astronomically solar. (0, Troll)

cheekyboy (598084) | more than 5 years ago | (#29395075)

And it will cost 50x the cost of cheap PCs?
Surely as google does it, its better to have 20 x cheap pcs each running E5400's or anything thats less than $100/cpu with $50 mb's

Everything eventually fails, its not worth spending 50x more for something that lasts 2x longer, when replacement costs are super
cheap, and the replacements are going to be even faster.

Re:And costs are astronomically solar. (4, Insightful)

Yvan256 (722131) | more than 5 years ago | (#29395117)

Google goes for the lowest watt per processing, the actual hardware cost is probably negligible compared to the cost of years of electricity for powering the systems and cooling the surroundings.

Re:And costs are astronomically solar. (0)

Anonymous Coward | more than 5 years ago | (#29396829)

Google goes for the lowest watt per processing

No they don't, if they did all their machines would be solar powered 400MHz ARM systems.

Google goes for the cheapest overall setup, just like everyone else.

Re:And costs are astronomically solar. (1, Insightful)

Anonymous Coward | more than 5 years ago | (#29395219)

And it will cost 50x the cost of cheap PCs?
Surely as google does it, its better to have 20 x cheap pcs each running E5400's or anything thats less than $100/cpu with $50 mb's

Everything eventually fails, its not worth spending 50x more for something that lasts 2x longer, when replacement costs are super
cheap, and the replacements are going to be even faster.

Google? How many tens or maybe hundreds of millions of dollars has Google spent developing the software that can run on piece-of-shit boxes?

It sure as hell is relevant to be able to buy one box that simple non-redundant apps can run on when the alternative is trying to pay massive amounts to develop fault-tolerant and redundant custom apps that can run on two or three cheap boxes.

Because unless you can run your software on lots and lots of boxes like Google does, it's cheaper to throw high-end hardware at availability and redundancy than it is to develop redundant and highly-available software. $50,000 gets you just a few months of developer time when you factor in all the overhead costs.

Re:And costs are astronomically solar. (1)

TheRaven64 (641858) | more than 5 years ago | (#29397171)

Google? How many tens or maybe hundreds of millions of dollars has Google spent developing the software that can run on piece-of-shit boxes?

Not many, but then Google has a problem that is naturally parallel with very few data dependencies. Not all of us are so fortunate.

Re:And costs are astronomically solar. (1)

raftpeople (844215) | more than 5 years ago | (#29399301)

"Not many": Do you know that? Because you posted that like you do actually know, as opposed to pure speculation.

My guess (just an opinion) is that it would absolutely require a large investment in custom software and manpower to create the infrastructure that google has created. And it makes sense for them and is probably worth it.

Re:And costs are astronomically solar. (1)

leenks (906881) | more than 5 years ago | (#29396601)

Clusters of machines aren't good for all problems, and Google doesn't use their huge clusters for everything... Some problems can only really be solved on a larger machine.

Hardware costs aren't the be all and end all any more either - one of the biggest costs is electricity (both for the machines and the cooling). It may well be that it is cheaper in the long run to have less of these than a huge cluster.

Anyway, to me it is quite clear why Oracle want Sparc and Solaris - have a good look at the Oracle product line, especially at products such as Exadata.

Leak to make Oracle look good? (1, Interesting)

Anonymous Coward | more than 5 years ago | (#29395093)

How'd they get this roadmap? More than likely from someone inside Oracle. Now when Oracle gets Sun and the SPARC chips are better than this, Oracle will get the credit for "saving Sun".

Or am I too cynical?

Re:Leak to make Oracle look good? (0)

Anonymous Coward | more than 5 years ago | (#29395313)

Actually, this doesn't make Oracle look good. It's just more of the same from Sun. Too little, too late, too expensive, and the wrong product. Oh... and it is just promises.

Hey, and I'm a SPARC/Solaris fan!

Re:Leak to make Oracle look good? (0)

Anonymous Coward | more than 5 years ago | (#29395583)

I'm a Solaris fan but SPARC you can keep. The specs are fine for what they are but there are fine 64 bit four and eight core servers out there that have similar real-world throughput from the likes of HP and IBM. They haven't been the dot in dot-com in a long, long time. Linux can do the job in lots of cases where Solaris would be a choice but the OS would be supported by Redhat or Suse. SPARC as a choice left when Sun sold to Oracle. Oracle isn't in the hardware business and is not likely to be in the hardware business six months from now.

Re:Leak to make Oracle look good? (1)

ToasterMonkey (467067) | more than 5 years ago | (#29403429)

Also a SPARC/Solaris fan... SUN really needs to polish their stuff up. They have a ton of cool tech, but it's very inaccessible to the average administrator. I consider myself to be fairly smart, and willing to go great lengths to learn new things, but I'm not who they need to market to. I wish they looked at the big picture and made their technology easy to use with fewer gotchas. The problem with Solaris.. I can't believe I'm saying this.. is UNIX. It really needs to go. I'm not talking about throwing out CLI's, KISS, openness, reuse, etc.. Just reevaluate what servers are used for today and design around that. Computers should be so damned easy to use today... when few have to manage many, the smallest usability improvement is multiplied, a LOT.

For example.. patching. Every other vendor can do this quickly and painfully now. I don't give a rat's ass how you implement it SUN, make it not suck. Backing out patches is a real cool feature, and slow, but Solaris is the only system I've ever had the need to revert an update. If it means we can no longer install hand stripped, custom slim builds, so f'ing be it. All full OEM instal then if thats required for decent QA. Linux can improve in this area too.. I'm sick and tired of all the minefields with "enter at own risk" signs. It is not necessary to be competitive. You need to look at the big picture, decide what users want, and guide them there. This is how Microsoft & Apple are eating everyone's lunches. Being "allowed" to walk the minefield ourselves it not a feature, it's just cheap.

I still love Sun's tech (and admire Linux), but the landmines are tiring.

Doesn't mean they'll build them (2, Insightful)

sunderland56 (621843) | more than 5 years ago | (#29395185)

Sun had a 486i workstation roadmap, too. They never built a single one. Marketing dreams on a PowerPoint slide doesn't mean squat.

Re:Sun's track record for chip delivery... (0)

Anonymous Coward | more than 5 years ago | (#29395293)

is not good. They make lots of promises and just don't deliver. SPARC is RIP... too bad but I just don't see it doing anything but losing market share.

x86 gives Windows, Linux, and Solaris flexibility, all with decent performance. Plus there is VMWare... It is really hard to make a case for further investment in SPARC in the data centre.

The T series are still wonderful machines but SPARC still looks like a dead end.

The SPARC VII chips are expensive, live in big ugly boxes (power hungry, space hungry) and do not deliver a performance premium over Intel and IBM.

I don't see how Oracle (a s/w company) will save a h/w company.
 

Re:Sun's track record for chip delivery... (2, Interesting)

mevets (322601) | more than 5 years ago | (#29395413)

On the first day I worked for Sun, an x86 implemented emulator of a SPARC ran at 2-5 % the speed of the SPARC of the day. Within 4 years, the same emulator on x86 of the day ran at about 50% the speed of SPARC of the day. By now, it may well be 1.2 times SPARC of the day. That is the nature of commodity electronics.

The story isn't SPARC vs x86; it is low-run specialized electronics vs huge commodity production. There is no craftsman-like advantage to the low-run specialty; this isn't furniture, it is etchings on polished sand.

Re:Sun's track record for chip delivery... (2, Interesting)

afidel (530433) | more than 5 years ago | (#29395795)

IBM seems to do ok keeping POWER up with and ahead of Intel/AMD. Sun just rested on their laurels and kept selling essentially the same very expensive product as the rest of the computer world was rapidly advancing around them. The only truly cool thing I saw come out of Sun was the T2 and Oracle killed it for me when they dropped the per core price break when T2 started getting too fast. Perhaps now that they own Sun the pricing will go back to reasonable but I'm certainly not holding my breath.

Re:Sun's track record for chip delivery... (1)

CharlyFoxtrot (1607527) | more than 5 years ago | (#29396433)

IBM seems to do ok keeping POWER up with and ahead of Intel/AMD.

Yeah but Sun never was IBM even though they seemed to think so themselves. IBM has a much broader range of products so they can prop up other business lines for a while if they have too. What's more they can sell during a downturn (buying IBM always being the safe bet) which is something Sun never could. Sun is very much a "boom" company, every time there's an economic slump they collapse like a bad souffle.

Re:Sun's track record for chip delivery... (2, Interesting)

TheRaven64 (641858) | more than 5 years ago | (#29397181)

IBM seems to do ok keeping POWER up with and ahead of Intel/AMD

Depends on your metric. Raw performance? Sure. Performance per Watt? Maybe. Performance per dollar? Not for most workloads. And don't forget that IBM's latest POWER chips use the same execution engines as the SystemZ CPUs, just with different instruction decoders and a few specialised parts unique to each design; the majority of both chips is the same. When you have customers who think $1m is cheap for a machine, this helps subsidise your workstation processors.

Re:Sun's track record for chip delivery... (1)

drinkypoo (153816) | more than 5 years ago | (#29397455)

Does it really matter how they do it? Their bigger problem is going to be finding commodity customers if they don't continue providing processors for all the consoles. The low end brings them into economies of scale that permit them to upgrade fabs etc.

Re:Sun's track record for chip delivery... (1)

chez69 (135760) | more than 5 years ago | (#29403083)

The big power boxes are really nice for a set of problems. we have a couple of big ass ones for oracle servers for very large databases and they work great. They do cost big $, as somebody said in a thread long ago when your data matters you get what you pay for. IBM makes a killing on these and the 'frames because for a lot of businesses it's easier to pay some big bucks now then later when your data is fucked. It's not as hard as you think to sell these guys, plus the virtualization is top noch, and done at the hardware level (I know you get that on wintel, I'm not sure if VMWARE the most popular solution uses it. I think sun also provides hardware assist for virtualization.)

That being said, for a lot of bulk work, I prefer linux/windows. Lots of speed on the cheap, perfect for a farm of app servers.

Re:Sun's track record for chip delivery... (1)

raftpeople (844215) | more than 5 years ago | (#29400177)

And don't forget that IBM's latest POWER chips use the same execution engines as the SystemZ CPUs, just with different instruction decoders and a few specialised parts unique to each design; the majority of both chips is the same.

And why do you think that is? Because they have been adding the mainframe ops to Power every generation so they can eventually have 1 single proc line instead of 2 (used to be 3 for servers, 4 if you include the mac stuff).

What is your point with this statement? That having all of the advanced features found on the mainframe CPU is a bad thing?

Re:Sun's track record for chip delivery... (1)

afidel (530433) | more than 5 years ago | (#29400575)

It doesn't matter, Sun was probably NEVER competitive on a performance per $ basis, but that wasn't their target market. You bought a Sun system because you needed a large, stable, high performance system with good industry support. Sun used to be all of those, but they pissed it away through mismanagement and poor execution. Almost all AIX customer are potential Solaris customers so if IBM can sell them expensive but performant kit then Sun should be able to as well if they are operating well but they haven't been for quite some time, which I find sad as I am actually a bit of a fan of Solaris.

Re:Doesn't mean they'll build them (1)

fm6 (162816) | more than 5 years ago | (#29396697)

Sun employees are not allowed to use Powerpoint.

Sun is DONE!!!! (0)

Anonymous Coward | more than 5 years ago | (#29395231)

Stick a fork in 'em!

CPU multi-threading (1)

ThePhilips (752041) | more than 5 years ago | (#29395327)

Can anybody give real life examples where the CPU multi-threading brings anything?

And please only real life examples: no theory, no official PR - I know them well myself.

Re:CPU multi-threading (2, Informative)

Nom du Keyboard (633989) | more than 5 years ago | (#29395407)

Can anybody give real life examples where the CPU multi-threading brings anything?

Multi-threading per core helps with video encoding. I saw benchmarks just today at http://www.anandtech.com/weblog/showpost.aspx?i=642 [anandtech.com] showing the results of the same processors run against the same tasks with and without HT enabled. How many thousand more examples do you need to see?

Re:CPU multi-threading (1)

ThePhilips (752041) | more than 5 years ago | (#29397111)

Thanks!

Re:CPU multi-threading (1)

goofy183 (451746) | more than 5 years ago | (#29395537)

How about any sort of web-server type task. I do development on web-based portal software that is highly threaded. Each thread doesn't due a huge amount of work but there are a lot of them (multiple threads per web server request) so having a machine that can run 128 threads (though each is fairly slow) easily outperforms a machine with much faster CPUs but only 4 or 8 of them.

Generally webserver type loads do better on hardware/clusters that can deal with lots of threads even if they aren't all that fast.

Re:CPU multi-threading (0)

Anonymous Coward | more than 5 years ago | (#29395617)

I used a t2000 (1.2 GHz), the T1 chip (8core x 4thread), for testing a SOA application. It was put up against HP BL25p (+3 GHz HT) servers running RHEL 4. The one t2000 outperformed the slightly older x86 servers by serving 17.75x the number of transactions. That server probably cost about 50% more than the two HPs.

Re:CPU multi-threading (1)

ThePhilips (752041) | more than 5 years ago | (#29397135)

SunFire vs. blade? Quite unfair comparison.

Re:CPU multi-threading (0)

Anonymous Coward | more than 5 years ago | (#29397451)

The chassis is irrelevant. You'd get the same performance out of a 32 thread T1 in a blade configuration.

Re:CPU multi-threading (1)

postmortem (906676) | more than 5 years ago | (#29395883)

If there was no multi threading, your IE/Firefox would be frozen until it completely loads any webpage.

Re:CPU multi-threading (0)

Anonymous Coward | more than 5 years ago | (#29395921)

Own a Xbox 360 or PS3? Games are designed to take advantage of the hardware threading and it is the difference between more simple environments with clunky AI and dynamic environments with thoughtful AI.

Just Give It Up Now (0, Flamebait)

Nom du Keyboard (633989) | more than 5 years ago | (#29395357)

Sun, if this is the best you can do -- 4 cores, 8 threads, arriving at 45nm just as everyone else is getting to 32nm -- just give it up now instead of asking us to watch a slow, agonizing, death.

Re:Just Give It Up Now (5, Interesting)

Daniel Phillips (238627) | more than 5 years ago | (#29395473)

Sun, if this is the best you can do -- 4 cores, 8 threads, arriving at 45nm just as everyone else is getting to 32nm

Sun's performance as a chip vendor is far better that your performance as a Slashdot troll. According to Sun's roadmap, a 16 core times 8 threads processor (128 threads just to be clear) at 40 nanometers arrives in 2010. That would be four sockets per blade, 48 blades per chassis for a respectable 768 multithreaded processors per chassis. As Sun says, it comes down to the TPC-C numbers. I'm no Sun fanboi, far from it, but I could be convinced by the right performance/heat ratio.

Re:Just Give It Up Now (0)

Anonymous Coward | more than 5 years ago | (#29400637)

Beckton will eat Sun alive in the 4 processor and below market. Sun's in deep shit. They have Itanium and Power to compete with at the high end, and Nehalem is moving its way up into the big leagues with Beckton. How about 32 _fast_ cores per blade, 64 _fast_ threads per blade? It will eat Sun's lunch in almost every conceivable way.

Re:Just Give It Up Now (1)

the linux geek (799780) | more than 5 years ago | (#29396179)

They're saying its 128 threads per chip, not eight, and at 40nm. Are you illiterate?

Re:Just Give It Up Now (1)

HuguesT (84078) | more than 5 years ago | (#29397137)

Sorry if this redundant, but the title is not very clear, we are talking about 8 threads *per core*, not 8 threads total like with the Intel i7.

SPARC Roadmap? (0)

Anonymous Coward | more than 5 years ago | (#29395677)

At this point, it looks pretty much the same as a (DEC) Alpha or Itanium roadmap.

Re:SPARC Roadmap? (2, Interesting)

Daniel Phillips (238627) | more than 5 years ago | (#29395755)

At this point, it looks pretty much the same as a (DEC) Alpha or Itanium roadmap.

Except for one thing: the SPARC circuitry is entirely open source [opensparc.net] . This has interesting implications, such as the fact that enthusiasts can build these things as FPGAs or even ASICs as fab costs come down to within the reach of clubs and schools. And emerging economies can fab these things by the bazillions without paying royalties. Not to mention big rich countries too. [theregister.co.uk]

Re:SPARC Roadmap? (1)

TheRaven64 (641858) | more than 5 years ago | (#29397199)

Sun have done a lot to encourage this. They make an OpenSPARC developer board, which comes with a Xilinix FPGA and peripherals and is basically a SPARC workstation with a CPU you can redesign in software. If you're a university, they'll send you some for free. If not, they'll send you some in exchange for a lot of money. There's also a book available as a PDF download from the OpenSPARC site describing the internals of the T1 and T2, which is an interesting read. I don't know if any contributions from outside Sun have made their way into the next generation OpenSPARC chips, but I wouldn't be surprised if some did in future.

Re:SPARC Roadmap? (1)

tuxicle (996538) | more than 5 years ago | (#29403261)

Here's the platform [digilentinc.com] that enthusiasts can use to tinker with OpenSPARC.

Ultra 27- two PCIe but no video cards (1)

purduephotog (218304) | more than 5 years ago | (#29397069)

The Ultra 27 was released with 2 PCIe x16 slots .... and it wasn't until we'd bought the damn things that we found out you can't put two FX-5600s in there- the case was designed to prevent it.

What's that got to do with their SPARC roadmap? Next x86 box we buy will be intel reference design. It's cheaper.

(not to mention there are bugs with the XVR-300 and the FX series of cards where you can't turn on 3 heads- it's 2 or 4 only)

Virtualisation missing (1)

markus_baertschi (259069) | more than 5 years ago | (#29397645)

I'm missing indications about better virtualisation features, like I'm used to them on IBM gear. These days all high end installation I see are running tens to hundreds of virtual machines on a single server. It looks to me that virtualisation in this scale is not even on the roadmap.

Markus

Re:Virtualisation missing (1)

Lally Singh (3427) | more than 5 years ago | (#29400519)

Solaris Zones not doing it for you?

Re:Virtualisation missing (1)

markus_baertschi (259069) | more than 5 years ago | (#29403827)

Solaris Zones, as I understand them, isolate applications from each other, but all are running within/on top of the same Solaris instance. As soon as you want to run different OS levels for the different apps or environments you are out of luck.

For example a new OS maintenance level is usually tested for a while in a test environment before being applied in production. Zones don't help here.

Often we have also incompatible prerequisite requirements of different apps (3rd party apps are terrible in this respect). App1 need at least maintenance level 123, while app2 has not been tested yet on this level and is not supported. If you give each a separate OS image, then you can give everyone what he wants.

You pay a small hardware price: A OS disk (30G) and some memory (512M), but you remove a ton of versioning update scheduling constraints.

Markus

Load More Comments
Slashdot Login

Need an Account?

Forgot your password?