Slashdot Mirror
Sun Considers Opteron
Sanjay writes "Official from Sun spokesman. Sun is considering using AMD's Opteron chip in a server it expects to deliver to the market shortly.
Intead of fighting Win of Wintel (like Redhat is doing), Sun can choose to fight both with Linux AMD's servers and also fight with HP/IBM as Itanium is anyway a non starter.
Sun can rise again!
"
and Slashdot has covered CNET's story
Will I retire or break 10K?
Here
:)
Whatever happened to those of us with acess to TMF being able to submit notice for pending dupes? I tried, but there's no easy way to figure out how to send a note to the editors. I still like the idea (naturally, since I brought it up) of a little form on TMF stories with the ability to submit dupe notification right then and there.
Of course, if I'm wrong, then, fine.
When I read "Sun can rise again" my wee mind read it as if Apu was saying it ala "Thank you, come again"
Trolling is a art,
And maybe we should change that slogan, what about:
News for the amnesiac. Stuff that mattered
After hearing that Microsoft is going to use it, and Now, Sun.
So when is Intel going to use Opteron?
SCO to Hell
Is taco really a human? The /. editor turing test (/.ETT) has been applied and the answer is ....
YES
Since any computer could be programmed to check for dupes. Unless of course TacoAI is sooo devious that it intentionally posts dupes to make is _seem_ like there is a real human being.
When Taco starts singing "Daisy", then we'll know the truth.
Let me get a press release together, hang on...
Is this truly the only Earth I can live on?
... It's a new story, and one that actually confirms that they ARE using AMD for something. The first one was saying 'don't count on it, but it might happen'.
So it's only a dupe in general topic, but if that's a true dupe, then everything that says 'New hole found in MS software' should also be a dupe.
----- - The beatings will continue until morale improves
SunMicroDevices monopoly, here I come!
How about a feature of SlashCode that, at preview time, it searches the site for words similar to ones in the story to be submitted, and displays the results. This would allow the submitter to determine if he is submitting a likely dupe.
This can be repeated for the poster...
April 8: Sun may use Opteron
April 9: Microsoft commits to Opteron
April 10: Sun considers Opteron
Who's following who ?
The problem for Sun is that Linux on Opteron does not give Sun much in the way of profits because the profit margin is low and competition is fierce. Sun cannot compete against IBM and HP in this area. Worse, Sun has no services organization to make any money by helping its customers to use Linux on Opteron.
Anyhow remember that stupid comment by Scott McNealy, who claimed that Sun is a one system -- one OS and one processor -- company. Now, Sun is distributing 2 OSes and 2 processors. Read the article at the Economist web site . It says that Sun will lose out big time in the Linux marketplace.
The Sun is setting. Good Riddance.
Is Sun going to Opteron a backdoor linux strategy in case hell freezes over and Sun decides to drop their Solaris all-together or straddle Solaris/Linux (again).
0 11217.shtml ?tid=1426
With Linus saying he really likes 64-bit strategy of Opteron vs. Itanic, perhaps they want to keep their options open. See these articles:
http://slashdot.org/articles/03/02/25/
http://www.theinquirer.net/?article=796
You've GOT to be kidding? If you were running desktops or small workstations, maybe. But, servers?!
Ever hot swap a CPU on a SMP PC? How about adding a CPU or RAM module without powering down? Hot sawp PCI? How about 4-way machines scalable to 64-way? 64+ Gb of RAM? Terabytes of storage?
PCs are only starting to be able to compete in that market, which is why Sun, IBM, and HP still sell those types of machines.
If you don't need those types of options, then PCs are fine.
Learning HOW to think is more important than learning WHAT to think.
Sanjay writes "Official from Slashdot spokesman. Slashdot is considering using Slashdot's Slashdot chip in a server it expects to deliver to the market shortly. Intead of fighting Slash of Slashdot (like Slashdot is doing), Slashdot can choose to fight both with Slashdot's servers and also fight with /. as Slashdot is anyway a non starter. Slashdot can rise again! "
Well...replacing most all of the proper nouns with Slashdot at least gave me a chuckle. Okay, so I'm retarded.
-AS
I'm glad this news *finally* makes it to slashdot as it certainly has *never* been discussed before. On a related note, has anyone heard about that new 'evil bit' RFC? I'm surprised that has never made it on slashdot.
XML is like violence. If it doesn't solve the problem, use more.
From the office of Iraqi Information Minister Mohammed Saeed al-Sahhaf (aka Baghdad Bob):
"Sun actually can rise again!???"
Think about it.
Karma: The shiznight, mostly because I am the Drizzle.
Sun needs to start looking at implementing new techniques in regards to it's OS and hardware integration business. Everyone knows that Sun UNIX can perform admirably and is very powerful, but at the same time, IBM is showing that Linux can be substituted for UNIX in the low to mid-level range, and HP has proven that they can sell Linux servers in the absence of huge corporate support.
Many pointy hairs are also awakening to the fact that Linux is evolving way faster then any previous OS in history. This realization is forcing many of them to position themselves in order to benefit from Linux. They are starting by replacing all of their low to medium-level extremely expensive UNIX solutions with Linux implementations, and waiting for Linux to overtake UNIX on the top tier. This saves them tons "in the meantime" and prepares them for the eventual replacement of their high-end solutions. Sun has to know that this scenario is inevitable and play along. Pride will only get you but so far.
McNealy has been fighting Linux for far too long, calling it "just another tool". I got news for you, all OS's are tools. Only this tool here can save your ass a ton while doing everything that every other tool promises to do on the low and medium ends.
Right now, Linux is "it" - and it shows no signs of slowing up. Microsoft makes their money off desktops and their office suite. UNIX makes money off stability and power. Stability and power is what the open source developers aim to improve. UNIX beware - evolve or perish, because you're next..
I really don't know what Slashdot editors do, but if they're not reading the site on a daily basis, couldn't they at least search the damn site before they post to see if someone has beaten them to it?
It's getting pretty rediculous. It wasn't always this bad.
Sun plans to use them in inexpensive blade servers which means they aren't totally abandoning RISC servers.
You're parked in the dark alone with your girl when she suddenly introduces you to Tammy and Buffy. "Girls Who Name Their Breasts" on the next Geraldo.
Given that so many companies (e.g., Sun, IBM, and Dell) want to increase their 64-bit x86 offerings, Microsoft had better work double time to release 64-bit Windows.
Imagine the horror of loading 64-bit Windows up on a Sun-branded server. Would the server implode in some sort of confused fury? Would the little OpenBoot PROM chip see the coming plauge, crawl out of the enclosure, and run away to a blissful place where that MS monster can't go?
Healthcare article at Kuro5hin
You apparently intend your post as a warning to Sun that it should give up on its existing technologies (sparc,solaris) and join what you perceive as the "linux pack" of IBM, HP, etc. But if you look at your own arguments and reconsider them, the case is far from clear that what you suggest is in fact wise.
Consider IBM. Sure, IBM is selling hardware with Linux loaded on it. But they haven't given up on their Power chips as you seem to imply that Sun should its Sparc series. Why aren't you wagging your tongue at IBM for that? And AIX? What of that? IBM certainly hasn't abandonded it, and I wouldn't expect it to any time soon. So all that IBM is really offering is yet another operating system choice, in this case Linux, and it meets your approval. It doesn't necessarily do anything unique there.
And what of HP? You say that HP has proven that they can sell Linux servers in the absence of huge corporate support. What on earth are you talking about? HP is one of the largest computer companies on the planet. If they can't make a go of selling Linux boxes, who can? I will also point out that HP hasn't dumped RISC for X86, but instead went to expensive Itanium, and has a long roadmap for HP/UX. Sure they will sell you a Linux box, but they would prefer to sell you something else.
You assert that Linux is evolving way faster then any previous OS in history. The only reason that is possible is because it has had so far to go to catch up. To catch up it has generally traveled trails blazed by others, and relied upon the kindness of volunteers and donations from kinfolk (JFS,XFS,etc.). Sure Linux is causing the traditional Unix vendors to react and jump a bit.... just like BSD did to AT&T Unix, GNU did to Unix, the various Unix groups and companies did to each other over time. But big unix companies are still here and adapting.
Cost? I've got Sun equipment that cheaper than my Dells, and suits my particular needs better. Cost/performance? Depending upon the day and the metric you've got a better argument. But it doesn't matter how cheap it is, or what the price/performance is if it doesn't cut the mustard. PC and linux aren't even close to being a universal solution. Check back in 3 years after Opteron is well entrenched, Linux gets some more time in the rock polisher, and companies have figured out which direction Linux on X86 is heading: Intel vs AMD. Till then, confusion reigns.
I also wouldn't count on Linux staying cheap. All of the major Linux commercial vendors are putting plans into place or releasing enterprise or professional releases that are both much more expensive, and have a much lower change rate. What else do you expect? Linux companies have been going broke left and right for years, and only a few now are starting to make a profit. There have to be profitable Linux vendors if linux is going to be a commercial success, and that means money, lots more money. And that money will come from their customers for license and support costs. I pay less for Sun support than what is in my budget for Red Hat support. It will be interesting to watch what happens to the Linux marketplace once that becomes more common.
The change rate for commercial linux is starting to drop for the professional releases. This has to happen since if you need something reliable that you are going to bet your business on, you can't afford the overhead of the constant release churning that has marked the Linux world to date. Testing, certification, and quality assurance take time. I wonder how that will effect Linux in the marketplace?
Its kind of ironic, but many of the things that you list as big advantages for Linux are really disadvantages to those with deep pockets. Rapid change is bad. Cheap is irrelevant. Almost as stable isn't stable. Those cost of the application, its implementation and maintenance is king. But the Linux commercial marketplace is heading toward those opportunities. I wonder what the outcome will be?
Linux is "
So I'm a pervert. Welcome to the Internet.
Sun has been hurting for a while - PC-based servers have been increasingly eating up Sun's market.
The Opterons are aimed squarely at a market segment that was hitherto tied to Sun and one or two other companies. If you wanted a highly-scalable 4- or 8-way 64-bit machine, you bent over, and Sun/IBM/DEC found your bank roll along the way.
Now, machines of those natures are coming from a commodity vendor. With a 128-bit DDR333 memory interface, each processer will have far more memory bandwidth than even the new Sun iiia's that were introduced today. And HyperThreading gives some pretty respectable inter-processor bandwidth. You think that Sun shouldn't be shaking in their boots? You bet they should.
In the end, they know that they're not going to win the lower end of the market. They simply can't compete with the economies of scale that AMD and Intel enjoy. Embracing the future is their only way to ensure that they keep at least a portion of that market.
steve
Oh, you're not stuck, you're just unable to let go of the onion rings.
If you don't see why it's a fight, then you don't understand economics!
Each company wants to succeed and make greater profits. In fact, they're *obligated* to their share holders to do just that. However, there's a finite market for these things, so they've all got to compete with each other - fighting.
Look at DEC's Alpha. A truly great product. But because the various companies that owned it weren't able to compete (fight) with it, look where it's at today.
steve
Oh, you're not stuck, you're just unable to let go of the onion rings.
Sun is now offering UltraSparc IIIi processors:
http://www.sun.com/processors/UltraSPARC-IIIi/
They do have some similarities to AMD's opteron processor:
- 1 MB on-chip L2 cache
- integrated memory controller
- 128bit DDR Ram
- large L1 cache
It should be interesting to compare those two processors.
Given how much time Sun has lost on the Linux revolution compared to rivals IBM, HP and even Dell, they need to make a concerted push in less than two directions.
I think the Solaris/x86 effort dilutes the strength of Sun's commitment to Linux. They can say that there's cross fertilization, but they're sending a mixed message to their customers. Those customers, like me, have appreciated Sun's UNIX experience, their leading the way with things like NFS, RPC, NIS and Java, and their emphasis on hardware reliability and performance.
Those customers are looking at the economics of Linux/x86 and like what they see. That's bad for Solaris/SPARC, except where the big iron hangs out. And the cut-off transition from where x86 won't suffice to mainframes that will do the job keeps moving up the food chain. Sun's food chain. The lucrative high end is becoming an ever shrinking market.
What does Linux need that Sun can do better than others?
Where Sun can make a big difference is in enterprise level management. Big directory/authentication services; interoperable services for managing heterogeneous LANS. Performance tuned next generation NAS/SAN services.
"Provided by the management for your protection."
I'm pretty sure Compaq and others already have hot-swap for cards, and support terabytes of storage (which is really just a matter of having enough FC bandwidth for whatever you're doing and plugging into the same standard storage arrays the Sun can).
Hot-swapping CPUs and RAM is trickier, but Sun only offers that on high end models which have no direct counterpart in the PC marketplace. Even then, it's a dicey situation at best.
With the E10K generation, you can hotswap CPU boards (there's 16 of them, each holding up to 4 processors, 4G ram, and two I/O busses (4x Sbus cards or 2x PCI cards). Thus you very much have to plan ahead to make sure you can "swap out" a given board without losing anything (oops, the failed memory is on the board with the only controller for this scsi disk over here, or the only one with this gigabit network connection). Assuming you built the machine right so that no single board is a single point of failure, you hit the next problem: If a CPU or memory module were to actually fail during runtime, it is still just as likely to cause an OS crash. The advantage is that in most cases the offending peice of hardware (1 CPU, 1 bank of RAM, etc) will be blacklisted and not used at all when the machine reboots from the panic (now you have a 15 CPU machine instead of 16). Then after that reboot, you can go about hot-swapping in a replacement with the OS online. You run some commands which basically tell the scheduler to stop scheduling on those CPUs, and tell the VM to not allocate any more physical ram in a certain region - then it goes about paging all the allocated RAM off to other ram or swap until it has emptied the board - then you can swap in the new stuff and re-add the CPU/mem into the OS.
On the newer SunFire architecture (3800s, 6800's, 15K, etc), they finally split the I/O boards from the CPU/Mem boards to make this a bit less painful, thank god. Still, in either case, you dont get a 4-way that scales to 64. You could buy a 64-capable machine (or higher now with SunFire architecture), and only populate it with 4 CPUs because you expect growth - but an E10K with just 4 CPUs in is a huge waste of cash - we're talking at least several hundred thousand dollars, for the hardware equivalent of what other companies sell for just a few thousand dollars. I think at one point a few years ago my company bought one 1/4 configured (16 CPU 16 GB ram) and left the other 3/4 open for expansion, and the cost was on the order of around $1,300,000. Do you really want to pay 50x+ over the same hardware capacity of a top end x86 just to be able to expand and have better support?
And in any case - these solutions, ultimately, may have slightly better sigma numbers on uptime, but they are still riddled with single points of failure, and ultimately no Sun solution is truly reliable with resorting to redundant clustering of oen sort or another. Once you resort to a redundant cluster, you're saying "I don't care if the hardware fails occasionally, my cluster will handle it while we do the maintenance". At that point, are you going to spend that much more money to make the difference between 99.9% and 99.999%?
Lets make a rough real example - a 24/7 Oracle database. In the Sun world, to get 24/7 uptime, you'd build out two machines of appropriate power (let's say 2x 6800s), and drop Oracle's OPS or RAC (or whatever they call the next generation) on it for a fully fault-tolerant cluster. You'd attach it to an FC SAN of appropriately configured redundant storage.
On the x86 side, you'd rack up the equivalent in I/O and CPU horsepower worth of 1U boxes (let's say 32x 1U dual processor large-ram crap-reliability boxes from Penguin Computing or something).
Either one is going to be very reliable because of Oracle's nonstop clustering stuff. You'll experience more failures/year on the x86 solution, but losing one of 32 machines is no biggie for a few hours while you drop in a spare.
Two fully loaded 6800's is gonna run you about $2.0 million. 32 high end-ish (lets say 10K a pop) 1U machines is gonna run you $0.32 million. You do the math.
11*43+456^2
Hell, I only spend a few minutes a day reading slashdot, and I have no trouble instantly spotting the dupes, so it wouldn't be too onerous a burden on your editors, would it?
Check out the SAP benchmark site: http://www.sap.com/benchmark/index.asp 1. Sun is the fastest computer available: Sun Fire 15000, 104-way SMP, UltraSPARC III, 1200 MHz, 8 MB L2 cache, score is 8000 2. If you divide the score by the amount of processors: 8000/104 = 76.9 compare with : IBM eServer xSeries 440 Model 8687-38X, 16-way SMP, Intel Xeon MP 2.0 GHz, 256 kB L2 cache, score 1090 score per cpu: 1090/16 = 68.1 So, even per-cpu basis, trivial UltraSparc III is faster that P4 Xeon 2.0 Ghz. Want to compare with Itanium? NEC Express5800, 1000series Highend Server, 32-way SMP, Itanium2, 1.0 GHz, 3MB L3 cache, score 2750. Per cpu: 2750/32 = 85 Ups! Super-duper Itanium with 3 (!!) Mb of cache is just a little bit faster.
Andrew
1. Sun is the fastest computer available:
Sun Fire 15000, 104-way SMP, UltraSPARC III, 1200 MHz, 8 MB L2 cache, score is 8000
2. If you divide the score by the amount of processors: 8000/104 = 76.9
compare with
IBM eServer xSeries 440 Model 8687-38X, 16-way SMP, Intel Xeon MP 2.0 GHz, 256 kB L2 cache, score 1090
score per cpu: 1090/16 = 68.1
So, even per-cpu basis, trivial UltraSparc III is faster that P4 Xeon 2.0 Ghz.
Want to compare with Itanium?
NEC Express5800, 1000series Highend Server, 32-way SMP, Itanium2, 1.0 GHz, 3MB L3 cache, score 2750.
Per cpu: 2750/32 = 85
Ups! Super-duper Itanium with 3 (!!) Mb of cache is just a little bit faster.
So, you need to consult real benchmarks, SPECint is just interesting exersise, which should not be taking into account for anything above 2-way servers.
Andrew
For those of you bitching that sun4u sucks and yada yada, I'm willing to bet you have never seen anything bigger than an Ultra 10.
Sun Fires are massive boxes. Will all the options that PC's could only dream about: System partitioning, Hot swap _everything_, killer backplane speeds (quad-port fast ethernet cards anyone?)..
True the lone UltraSPARC processor is fairly unimpressive, but in an E12K you can have up to 256 of them if I recall. That's on one single, operating system. So take your silly 48-node Athlon clusters and go home.
Just trying to come to the defense of an arch that really isn't bad when you're not trying to run Lunix on it and play games with WineX.
I just wasted your mod points! HA!
"Assuming you built the machine right so that no single board is a single point of failure, you hit the next problem: If a CPU or memory module were to actually fail during runtime, it is still just as likely to cause an OS crash. "
m in ata_10-11-02.pdf
4 35/spainho wer.html
Not if you buy a mainframe class system from IBM or the other genuine high end vendors, which have things like redundant CPUs running the same code. If there's an error, the CPUs retry. If still bad, then the offending CPU or module is shutdown.
Consider Fujitsu if you still like SPARC, but want stuff like instruction retry.
http://www.ftsi.fujitsu.com/services/press/illu
Look at the IBM mainframe culture and history:
http://www.research.ibm.com/journal/rd/
Sun is a mainframe wannabe with decent marketing. They really aren't that far ahead of Dell if you look at the big picture.
Sun SPARC is actually lagging behind Fujitsu SPARC in performance and reliability.
Not saying Sun is dead or dying. But it doesn't look good does it?