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!

AMD Gains In the TOP500 List

timothy posted more than 3 years ago | from the included-chart-awesome dept.

AMD 77

MojoKid writes "AMD recently announced its share of the TOP500 supercomputer list has grown 15 percent in the past six months. The company credits industry trends, upgrade paths, and competitive pricing for the increase. Of the 68 Opteron-based systems on the list, more than half of them use the Opteron 6100 series processors. The inflection point was marked by AMD's launch of their Magny-Cours architecture more than a year ago and includes the twelve-core Opteron 6180 SE at 2.5GHz at one end and two low-power parts at the other. Magny-Cours adoption is important. Companies typically don't upgrade HPC clusters with new CPUs, but AMD is billing their next-gen Interlagos architecture as a drop-in option for Magny-Cours. As such, it'll offer up to 2x the cores as well as equal-to or faster clock speeds."

cancel ×

77 comments

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

So? (1)

Anonymous Coward | more than 3 years ago | (#36579830)

Why is it newsworthy when one company goes up or down a random list?

I guess if you are a stockholder, but other than that I can't see caring.

Re:So? (1)

Anonymous Coward | more than 3 years ago | (#36579910)

Why is it newsworthy when one company goes up or down a random list?

It isn't.

Re:So? (2, Insightful)

zill (1690130) | more than 3 years ago | (#36579942)

This is newsworthy because it's a new cache of ammunition for the fanboys. And since this is slashdot, we're basically at ground zero of the thermonuclear flame war.

Re:So? (1)

aliquis (678370) | more than 3 years ago | (#36580150)

What is there to fight about?

I won't flame but yeah.. Pick your product category and yada yada ;D

Re:So? (3, Funny)

Anonymous Coward | more than 3 years ago | (#36580182)

What is there to fight about?

A valid question! We know nothing about them, their language, their history or what they look like. But we can assume this. They stand for everything we don't stand for. Also they told me you guys look like dorks.

Re:So? (0)

Anonymous Coward | more than 3 years ago | (#36580236)

>=( THEY LOOK LIKE DORKS!

Re:So? (0)

Anonymous Coward | more than 3 years ago | (#36589390)

THEY look like dorks!! GRRR!!!

Re:So? (4, Interesting)

hairyfeet (841228) | more than 3 years ago | (#36580264)

Well I'd say its also news as many of the guys here are in charge of purchases and this shows AMD still has some competitive server offerings, and as someone who remembers what it was like when there was only Intel (crazy ass pricing, lousy chips, insane motherboard costs, etc) having competition is VERY important!

I'd add these gains were done in spite of Intel doing everything but setting the AMD factories on fire trying to kill AMD. They rigged their compiler (and still are BTW) to put out shit code if it runs on an AMD CPU, they bribed the OEMs so much that there were several quarters where the ONLY profits Dell saw were Intel kickbacks, and Toshiba said their kickbacks were so generous they were "like cocaine" to the OEMs.

So I'd say that AMD making gains despite Intel doing everything but breaking their employee's kneecaps just shows they still have some competitive spirit. I personally have switched my shop to AMD only after finding out about the bribery and compiler rigging and my customers couldn't be happier. Their netbooks and laptops all have smooth video and are easy to hook to a TV via HDMI thanks to the Radeon IGPs, their desktops are likewise great with smooth 1080p video and hardware acceleration of all the major formats as well as hardware transcoding, their triples and quads make a great centerpiece for a good cheap media box or HTPC, all in all they make a great consumer box that will do all your average person will ever want to do at a price they can easily afford without breaking the bank.

So I'm personally glad for TFA, as it shows me they have competitive server chips to go with their excellent desktop and notebook lines. And frankly you should be happy too as real free market competition is good for everyone. or do you honestly think Intel would have a chip that costs less than $1000 if it weren't for AMD?

some proof would be nice (0)

ArchieBunker (132337) | more than 3 years ago | (#36580468)

Any proof of these claims of compiler tampering? So why would you be using an intel compiler on AMD cpus? Does AMD not write one? Your third paragraph reads like an advertisement,

"Their netbooks and laptops all have smooth video and are easy to hook to a TV via HDMI thanks to the Radeon IGPs, their desktops are likewise great with smooth 1080p video and hardware acceleration of all the major formats as well as hardware transcoding, their triples and quads make a great centerpiece for a good cheap media box or HTPC, all in all they make a great consumer box that will do all your average person will ever want to do at a price they can easily afford without breaking the bank."

Seriously who talks like that besides maybe an AMD employee?

Re:some proof would be nice (1)

Anonymous Coward | more than 3 years ago | (#36580578)

No, AMD does not write a C++ compiler. In anyu case, that's irrelevant, unless you think software developers should start distributing two versions of every executable.

And it's a well documented fact that Intel's compiler checks for "GenuineIntel" in the CPU ID string and executes different (generally less efficient) code if it gets "AuthenticAMD" instead. Intel itself has never denied it, and says it's because "they can't keep track of what instructions other manufacturers support" (it's not as if they copied their 64-bit extensions from AMD or anything *cough*).

http://patch-authenticamd.googlecode.com/hg/web/index.html?r=1.0
http://www.agner.org/optimize/blog/read.php?i=49
http://www.theinquirer.net/inquirer/news/1567108/intel-compiler-cripples-code-amd-via-chips
etc...

Re:some proof would be nice (1)

ArchieBunker (132337) | more than 3 years ago | (#36580676)

So why should intel have to write optimizations into their compiler for a competitor? The article basically says if it finds a non intel chip then all optimizations are turned off. Can this feature not be overidden? Sounds overblown if you ask me.

Re:some proof would be nice (1)

Anonymous Coward | more than 3 years ago | (#36581108)

It's not that Intel had work to do to support non-Intel chips. They did extra work to make all code produced by the Intel compiler slower that necessary on non-Intel chips. Thats' not just binaries delivered by Intel, but also a lot of code from 3rd parties.

Re:some proof would be nice (2)

Chris Burke (6130) | more than 3 years ago | (#36585060)

So why should intel have to write optimizations into their compiler for a competitor?

They don't. The issue isn't "optimizations" as one usually thinks of it, tweaking a piece of code with compiler tricks to make it run faster.

The issue is using entirely different code paths that use x86 extensions like the SSE family of instructions. These instructions, in addition to being SIMD for when the same operation is being performed on multiple pieces of data, are essentially a replacement for the old x87 floating point co-processor instructions. x87, which is old and stupid, and a right bitch for both programmers and hardware designers.

There is no processor which supports SSE where that codepath would not be faster than x87. Intel doesn't have to optimize the SSE code for AMD, and nobody is asking them to -- AMD optimizes their implementation to be fast with the code Intel (and other compilers) produce.

The problem comes when the code Intel produces specifically ignores their own spec on how to detect if these instructions are supported, and instead checks for a "GenuineIntel" vendor string, and resorts to the x87 codepath for anything else. Even though AMD chips support both the SSE instruction set and the specification for detecting it.

So that's the problem. Intel is deliberately de-optimizing competitor's parts.

Can the feature be overridden? Yeah. If you know about it, you can hack the binary to remove the check, or you could use a microcode patch to make an AMD chip return an Intel vendor string. The former is illegal if you are an end-user of a non-Free program, and the latter is illegal if you're AMD.

Is it overblown? Maybe. It's a direct and flagrant instance of Intel taking action not to boost themselves, but to harm a competitor. It costs AMD quite a bit of performance in a variety of common and popular programs, as would be expected, and which can be shown by disabling the "feature".

Personally I think that when the solution is so simple and logical -- Intel honors their own damn spec and the CPUID bits for SSE and other instruction sets -- not doing it represents just maliciousness on Intel's part, and so the level of reaction is warranted.

Re:some proof would be nice (1)

TheRaven64 (641858) | more than 3 years ago | (#36582784)

Intel's excuse is valid. AMD has been fairly good (although not perfect) at only setting CPUID feature flags when they have a 100% compatible implementation of an extension. Other x86 manufacturers have been very bad. Worse, some (e.g. IDT's WinChip) have advertised some features and then implemented them with trap-and-emulate code in a driver. They did this for some uncommon operations, to make code that used them occasionally run, rather than exiting with an error, that meant that a compiler that used those instructions in an optimisation would make code that ran much slower. Intel's compiler does not assume that non-Intel chips that advertise a feature actually support it, because the team would get into a lot more trouble if their code broke on other manufacturers' chips than if it just ran a bit slower.

Re:some proof would be nice (1)

TheRaven64 (641858) | more than 3 years ago | (#36582820)

I forgot to say - even if the non-Intel chip has a 100% compatible implementation of the feature, that does not mean that it will have the same performance characteristics. This is true even among Intel chips. Some SSE-based optimisations can give you a 2-4x speedup on Core chips, but give a slowdown on Atom chips (early ones, at least, not sure about the latest one). Doing the same optimisations for an AMD or Via chips may make things faster, or they may make things slower. Without extensive testing, Intel's compiler team doesn't know which, and doesn't bother spending money on testing their competitors' hardware to find out.

Re:some proof would be nice (0)

Anonymous Coward | more than 3 years ago | (#36580596)

The stories were all over the place a few months ago. Something about turning standard processor features off despite the cpu having the flags, if it didn't have an intel cpuid.

Many companies compile their software for distribution to the public with the intel compiler. AMD does make one.

The latter part is indeed suspect.

Re:some proof would be nice (0)

Anonymous Coward | more than 3 years ago | (#36581494)

No, AMD does not make a C++ compiler. They have contributed some optimizations to gcc, and they provide Intel with all the necessary documentation to let their copiler generate optimized code for AMD CPUs, but they do not develop a separate compiler.

Re:some proof would be nice (1)

rbrausse (1319883) | more than 3 years ago | (#36582154)

No, AMD does not make a C++ compiler.

what about Open64 [amd.com] ?

Re:some proof would be nice (1)

TheRaven64 (641858) | more than 3 years ago | (#36582706)

Open64 is not developed by AMD. They ship a few Open64-derived products, but they aren't the main developers.

Re:some proof would be nice (4, Informative)

godrik (1287354) | more than 3 years ago | (#36580742)

"Any proof of these claims of compiler tampering?"

This is a well known issue with the intel compiler which has been fixed since. The story is told on wikipedia in the criticism section of http://en.wikipedia.org/wiki/Intel_C%2B%2B_Compiler [wikipedia.org]

The problem is so well known, that people wrote software to patch the code produced by the intel compiler to make it work properly on AMD processors such as http://code.google.com/p/patch-authenticamd/ [google.com]

"So why would you be using an intel compiler on AMD cpus?"

One of the interest in using the x86 instruction set is to be binary compatible so that you can use the code generated by any compiler. The intel compiler is a very good compiler, why not use it ? VIA also produces x86 processors you can use the binaries generated by the intel compiler on it. These technologies are designed to be compatible.

"Does AMD not write one?"

AMD contributes to http://en.wikipedia.org/wiki/Open64 [wikipedia.org] and to GCC.

"Your third paragraph reads like an advertisement,"

I agree on that one.

Re:some proof would be nice (0)

Anonymous Coward | more than 3 years ago | (#36581260)

Any proof of these claims of compiler tampering?

L2Google. Also: http://yro.slashdot.org/comments.pl?sid=155593&cid=13042922 [slashdot.org]

Re:some proof would be nice (1)

hairyfeet (841228) | more than 3 years ago | (#36581446)

Would you have preferred it if I had wrote "Yo Dawg, I be hooking up video and that shit be fat yo! teh pic be all big and shit, and it be fatter than a ho's ass dawg!" is that more preferable? How else are you gonna write that you get great video performance and the ability to transcode without actually saying that?

So excuse me for actually writing down what I have experienced installing and building AMD systems. Perhaps a video of never gonna give you up or a pic of a distended anus would be more to your liking?

Now back to the topic, I have found since switching that the typical 10% performance gain of Intel usually comes at a 300% markup when you figure in the price of the boards, and most people will never be able to notice the difference. I bet if I built identical quads, one Intel and one AMD, and had you do your average everyday tasks you frankly wouldn't be able to tell the difference. Well not until you got the bill that is.

And frankly anyone that supports the free market would have to be nuts to buy Intel ATM, because their behavior frankly should have had their ass dragged into court for antitrust years ago. Bribery? Rigging software? Oh and BTW Intel still rigs their compiler the ONLY difference since the settlement is they document it in the fine print and give you the "choice" of taking a dump over ALL of the software you compile by forcing it all to run in X87 mode, or only have it take a shit on AMD and Via chips. Wow, what a choice!

When you add to this their current campaign to slowly kill Nvidia, whose chipset division they have already strangled, then how anyone here can say that MSFT deserved to be busted but Intel not I will simply not understand. So if you don't want to reward bribery, unfair market manipulation, and outright douchebaggery, try to buy AMD next time okay? And this is from someone that for nearly 15 years was strictly an Intel man, but I can't reward douchebaggery with a clear conscience.

And as for proof, since teh Google must not work where you are here [slashdot.org] you [betanews.com] Go [theinquirer.net] and you should read the first one especially, since it is a post from right here on /. where a programmer documents the behavior and what happened when he ran programs compiled with an Intel compiler on AMD. Now since most programs on Windows (including most benchmarking suites) and compiled for Windows with the Intel compiler than the results of ALL those programs must now be looked upon as suspect. It is no different than the classic "Quack.exe" where if it detects A you get one result and B gets you another, even though the chips in question have identical abilities to run SSE code. And this programmer had noticed the problem back in 2004 so for at LEAST 7 years they have been rigging.

Re:some proof would be nice (0)

Anonymous Coward | more than 3 years ago | (#36581842)

Now since most programs on Windows (including most benchmarking suites) and compiled for Windows with the Intel compiler than the results of ALL those programs must now be looked upon as suspect.

Only the benchmark results. The actual important results (the output) will be correct. Really Intel are just being safe here since unlike with their own hardware they are unable to guarantee how functions are implemented in AMD's hardware. I would rather that my results were correct and slow than that they were wrong because the compiler writer expected something to work which doesn't due to a bug they don't know about.

I must confess I don't do a lot of development on Windows (it's about as common in HPC as Linux is on the desktop) so I don't know what is the "normal" compiler to use there, but there are a lot of different compilers for Linux. Generally on sites with AMD processors other commercial compilers are used (e.g. PGI). There's certainly no requirement in HPC to use the Intel compiler and sites have to buy a commercial compiler to support Fortran users (because while GFortran has come leaps and bounds in the last few years most of the commercial compilers are better) anyway so I don't really see what the problem is: If you buy Intel chips you buy the Intel compilers, Power chips buy the IBM compilers and so on. Maybe the real problem here is that AMD have skimped by not providing a commercial compiler and are upset when they can't free-load on the Intel one.

Far more exciting to me than this Intel/AMD fanboy war is that a machine with SPARC is back on top of the list. Proper processors are back folks!

Re:some proof would be nice (1)

hairyfeet (841228) | more than 3 years ago | (#36585576)

Except that ISN'T what they are doing. If you bothered to read the post this all started back when the first P4s came out, aka little space heaters big pile o' suck. if you'll notice it also crippled performance on the P3 which was currently kicking the dog snot out of Netburst (and eventually came back under the Core moniker) so what we have is a simple "If chip equals what we want to promote then code=fast" and "If ship equals one we want to bomb then code' shit".

If they simply wanted to be safe then the safest thing to do would be to do nothing at all. After all if someone complains that just helps them sell chips, right? They could say "Sorry, please use genuine Intel next time" and that would be that. But no, that would let people know they are being fucked. Otherwise how do you explain disabling SSE support on the P3, which had an Intel designed SSE path? Can't be worries about compatibility, after all Intel designed the thing!

No if you read the results in the first link sadly there is only ONE conclusion, and that was Intel is using their position as leader in the market to hamstring any chips they did not want to promote. The P3 really is the smoking gun here, as at the time the benchmarks were showing the P3 on average 20% faster than the Netburst while using 40%+ less power. But guess what happened after the Intel compiler rigging? Why Netburst was suddenly 22% faster than the P3! Isn't that amazing? Oh and while there may be other compilers on Linux, on Windows it is Intel or MSFT. Kinda sad when MSFT is the fairer choice huh?

The simple fact is we aren't talking about some esoteric feature that may or may not be supported, we are talking about basic SSE which is a published standard and has been supported by ALL chip manufacturers since before the Intel compilers were rigged. No this is no different than if MSFT had OEMs rig the BIOS of laptops to do a check for "Genuine Windows" and have it run like shit if the flag wasn't tested. This check does NOTHING but look at CPUID, that's it. No checking of MXX, or SSE, or any other flag. It just looks for a P4 or above flag and if it isn't found drops the entire stack into slow ass x87 mode, which frankly has been depreciated since the Pentium 1 and is used by NOBODY else. If you read the first link the programmer points out they even used a poorly documented crap memory move just to ensure that it would REALLY slow it down.

This isn't "playing it safe" this is classic Sherman Antitrust behavior that sadly Intel WILL get away with, because our government was sold to the corps years ago. Hell if the MSFT case happened today they wouldn't even have had a trial, hell they would have probably gotten a tax break for doing it!

Re:So? (0)

Anonymous Coward | more than 3 years ago | (#36580686)

I'm glad that AMD exists, because they provide competition and thus push Intel to innovate and lower their prices. I'm also hopeful that ARM will enter the Server, Workstation, desktop and laptop market more strongly with their highly parallel low power CPUs, since all my work is fully distributed and able to take advantage of any number of given cores...

But I would not bet on GPU accelerated systems as the way to go, since you have to rewrite everything for that approach (CUDA, OpenCL...). Instead I'm betting on the chips like Knights Korner (48+ cores, cloud on a chip), Polaris (original prototype, 80 cores), and AMD's future chips with 12+ cores. For which we can use our standard languages like Erlang, OCaml, MPI, distributed Haskel, Go... without having to rewrite things in cumbersome APIs...

Re:So? (2)

KingMotley (944240) | more than 3 years ago | (#36580800)

I'm glad AMD is around, but seriously you don't need to exaggerate things to make a point, especially when the exaggeration is obviously false.

Yes, Intel would have chips that cost less than $1000 if AMD weren't around. They had them BEFORE AMD was around, and that hasn't changed. They quite often do have their highest speed desktop processor right around the $1000 price, with another running at 90% of it's speed at $600, and continuing down the line. I don't suspect that will be changing any time soon.

Re:So? (1)

TheRaven64 (641858) | more than 3 years ago | (#36582838)

They had them BEFORE AMD was around, and that hasn't changed

When was this period? AMD was set up because IBM refused to buy the 8088 from Intel without a second source. AMD has been around for as long as the PC architecture...

Re:So? (1)

WuphonsReach (684551) | more than 3 years ago | (#36593294)

Look back to the advent of dual-core CPUs in the mid-2000s.

Intel offerings were server-only chips, all well over $300. AMD offered something that did the same thing, had better performance, and broke the $200 barrier.

It wasn't until AMD started mopping the floor with their Opteron / Athlon64 X2 chips that Intel got its act together and released affordable multi-core offerings. And it took them even longer to jump on the 64bit wagon that AMD had been pulling for 2+ years prior.

Remember - Intel's vision of the future for 64bit computing was Itanium.

Re:So? (1)

markass530 (870112) | more than 3 years ago | (#36580484)

whats with all the bitching about stories? If you can't find something to discuss about the story presented - go play on 4chan

Re:So? (1)

Anonymous Coward | more than 3 years ago | (#36579980)

You must be new here.

Re:So? (0)

Anonymous Coward | more than 3 years ago | (#36580064)

Applying for that Slashdot censorship job again, are ya? I recommend you build your own news aggregating website, so you aren't forced to click & endure these terrible, terrible posts.

Random? (5, Insightful)

Anonymous Coward | more than 3 years ago | (#36580126)

That's hardly a "random list". And this is a site about technology. Do you think it would be irrelevant for a site about gaming consoles to mention that the Kinect increased the Xbox's share of exercise games (or whatever) ?

Supercomputing designs are a rather good real-world benchmark of CPUs, because they take into account cost, density, ease of deployment, etc..

More interesting than the share of the top 500 list, though, is the fact that AMD holds the #3 spot, ahead of Intel (whose first system comes in at #4) and both are behind a GPU-based system (using Nvidia Tesla chips, at #2). This is probably a good indicator of things to come, and explains why AMD is betting so heavily on integrating the GPU into its overall system architecture, and why Intel has finally started serious work on GPUs.

#1, by the way, is held by a SPARC system.

Re:Random? (1)

jd (1658) | more than 3 years ago | (#36580396)

It's random if you're not into news for nerds. :)

Re:Random? (1)

dwater (72834) | more than 3 years ago | (#36581210)

It's random as in if you pick a list at random, then you could pick that one.

Re:So? (1)

aliquis (678370) | more than 3 years ago | (#36580146)

It's about as newsworthy as "random guy got aspergers", but hey, atleast these two stories aren't duplicates of each other!

Re:So? (4, Informative)

MacTO (1161105) | more than 3 years ago | (#36580290)

Well, it is of interest to people who are interested in supercomputers.

People who develop software for those beasts like to know how things are changing. They may not need to know the intimate details since compilers and libraries will handle most of that, but they may want to throw together a small cluster to test emerging technologies. (This is particularly true in recent years since small clusters based upon AMD/Intel CPUs and AMD/Nvidia GPUs are within reach of individuals.)

Stockholders though couldn't care less. The number of units is too small and the prestige counts for nothing unless it translates into sales in other markets.

Re:So? (4, Insightful)

jd (1658) | more than 3 years ago | (#36580424)

It's a bit more. Supercomputers push bus speeds and floating-point calculations to the limits. Much of the hardware, these days, is commodity and these two criteria are important in gaming, computer art (such as rendering). The former is also important in embedded computing (such as phones) since data streaming is limited by maximum sustained bus speed.

Re:So? (1)

Kaptain Kruton (854928) | more than 3 years ago | (#36581568)

This sort of information can also be useful for sales people. Many people don't know anything about PCs when they purchase a home computer. Many customers have heard of Intel, but the number of people that have heard of AMD is much less and they don't know if their products are any good. If a sales person wants to sell a particular computer that will fit their needs, the customer may be hesitant to buy an AMD based system if they haven't heard of it... even if the sales person says it fits their needs perfectly. However, if the sales person can tell them give them some nice stats, such as "15% of the top super computers in the world use AMD chips" or "the fastest super computer in the US (and 3rd fastest in the world) used AMD CPUs", then the customer will understand they are a legitimate and respected brand.

Also, the fact that AMD's numbers are growing helps indicate that Intel is getting a little bit more competition.... which is usually good for us geeks, in terms of price and performance.

Congratulations (0)

aywang31 (2078102) | more than 3 years ago | (#36579972)

Congratulations, AMD. You have just added 9 more computers to the TOP500 list. A whopping 14% of the TOP500 list is now AMD!! I think AMD is still sore at Intel for it's Core 2 Duo and Core i3/5/7 processors.

Re:Congratulations (0)

Anonymous Coward | more than 3 years ago | (#36580040)

so the trolling begins

Re:Congratulations (3, Interesting)

mehemiah (971799) | more than 3 years ago | (#36580080)

funny, whats more news worthy is that a SPARC chip reached the top. That hasn't happened in like a decade has it? (i haven't checked it in a while)

Re:Congratulations (3, Informative)

the linux geek (799780) | more than 3 years ago | (#36580132)

It's not standard SPARC. The VIIIfx's commercial-workload performance would probably be pretty bad - the cores themselves are more or less identical to the existing VII cores, which have less than impressive performance. The VIIIfx derives its HPC performance, which is admittedly good, from some extensions (called HPC-ACE) that are not part of the normal SPARC instruction set. In a lot of ways, the VIIIfx is closer to a vector processor than a conventional SPARC chip. This isn't inherently a bad thing, but it is certainly not a general-purpose processor.

Re:Congratulations (0)

Anonymous Coward | more than 3 years ago | (#36580240)

Not so much "a SPARC chip", more like a cluster of 550 thousand modified SPARC chips (well, cores). That's more processors than #2 and #3 combined.

Re:Congratulations (3, Interesting)

wisty (1335733) | more than 3 years ago | (#36580196)

I think Intel, with its Larrabee thingy (now Knights Ferry?) will come to dominate the lists even more, in the short term. But honestly, who wants it?

People (like, 99% of people) want a reasonable CPU, and maybe a GPU for games. Most researchers want fast cores (parallel problems are hard), or vector processing (Matlab, Scipy, Mathematica etc take the pain out of distributing the workload), or distributed systems; in that order.

So normal people will want Fusion for its cheap GPU, and researchers will want it because it's actually easier to program.

AMD might have a bright medium-term future.

Highlights as per Top500 site (4, Interesting)

gupg (58086) | more than 3 years ago | (#36580168)

The top500 site has its own take on highlights:
http://www.top500.org/lists/2011/06/press-release [top500.org]

- The two Chinese systems at No. 2 and No. 4 and the Japanese Tsubame 2.0 system at No. 5 are all using NVIDIA GPUs to accelerate computation, and a total of 19 systems on the list are using GPU technology.
- China keeps increasing its number of systems and is now up to 62, making it clearly the No. 2 country as a user of HPC, ahead of Germany, UK, Japan and France.
- Intel continues to provide the processors for the largest share (77.4 percent) of TOP500 systems. Intel’s Westmere processors increased their presence in the list strongly with 169 systems, compared with 56 in the last list.
- Quad-core processors are used in 46.2 percent of the systems, while already 42.4 percent of the systems use processors with six or more cores.
- Cray defended the No. 2 spot in market share by total against Fujitsu, but IBM stays well ahead of either. Cray’s XT system series remains very popular for big research customers, with three systems in the TOP 10 (one new and two previously listed).

In my opinion, the newest & most important trend in high performance computing is the advent of accelerators like GPUs.

Well the other thing I'd say it shows (2)

Sycraft-fu (314770) | more than 3 years ago | (#36581190)

Is we need to look at finding a new benchmark perhaps for supercomputer. Maybe two or three different kinds, depending on the kinds of thing you want.

The reason is that while GPUs are great, they are limited. We don't use them for everything for good reason. So if you have the kind of problem GPUs are good at, and linpack happens to be one of them these days, then wonderful. They are a great way to go in terms of performance/dollar and performance/watt. However if you don't, then they are not useful and you need to use CPUs.

Thus there are reasons to have the "slower" CPU based supercomputers because for some things, they are not slower at all.

For example one problem linpack has is it doesn't test interconnect speed very well. You can break the problem down in to small pieces and run it on the nodes quite well. Now that's wonderful, but not all problems are like that. Some do heavy cross node memory access and linpack doesn't show that.

That is an area GPUs fall down in. For example a Tesla M2090 has 177GB/sec of memory bandwidth to its onboard 6GB of RAM. Nice. However it only has 8GB/sec of bandwidth back to any system memory since it is on a PCIe 2.1 16x slot and that is all the bandwidth that has. Well that means things slow down a TON if you have to go off of the included RAM, never mind to RAM in another system.

So for some things, and linpack is one, that you can divide down and not have a ton of access between nodes, then wonderful, the GPUs are good. But we should have a benchmark that tests other situations too. Part of the reason why someone may buy a supercomputer instead of just a cluster is the need for high interconnect speed.

Re:Well the other thing I'd say it shows (1)

gupg (58086) | more than 3 years ago | (#36590474)

I think this is a big misconception about GPUs. They are good at many applications - not just Linpack.

Take a look at the list of applications ranging from video transcoding to weather forecasting to computational chemistry to physics at:
http://www.nvidia.com/cuda [nvidia.com]

In fact, the researchers at the Chinese Academy of Sciences just ran one of the fastest scientific simulations using their GPU supercomputer (#2 on the Top500 list):
http://blogs.nvidia.com/2011/06/chinas-investment-in-gpu-supercomputing-begins-to-pay-off-big-time/ [nvidia.com]

There are tons of papers at the Supercomputing conference for real "full" applications in a very diverse range of applications that are accelerated using GPUs.

Re:Highlights as per Top500 site (0)

Anonymous Coward | more than 3 years ago | (#36582236)

Let's remove a little AMD spin from the announcement. AMD still accounts for less than 20% of the HPC market. The SPARC processor accounts for about 1% and the #1 spot, though it is SPARC, is based upon custom spun silicon (you cannot buy the chip from Oracle). Furthermore, AMD is not gaining share from Intel so much as taking share away from IBM's Power processors. Another important trend is the move to optimizing toward GPU use, which may make AMD's lower priced, power-efficient CPUs more attractive.

Re:Highlights as per Top500 site (1)

gumbi west (610122) | more than 3 years ago | (#36582724)

What happened to IBM's power processor?

Re:Highlights as per Top500 site (1)

TheRaven64 (641858) | more than 3 years ago | (#36582900)

though it is SPARC, is based upon custom spun silicon (you cannot buy the chip from Oracle)

Why would you go to Oracle to buy a Fujitsu chip? I mean, they'll sell you rebadged Fujitsu chips if you want (the current UltraSPARC line, with the exception of the Tx series, are Fujitsu SPARC64 chips), but you'd have better luck going to Fujitsu. They'll happily sell you SPARC64 VIIIfx chips, if you can afford them...

My Company Contributed to This (4, Informative)

friedmud (512466) | more than 3 years ago | (#36580178)

The supercomputer I use daily is one of these new AMD based ones in the TOP500. It is a sweet machine. My software (custom engineering simulation written in C++) scales perfectly on it all the way out to over 10,000 cores.

The memory architecture is really excellent as well. With our old Intel based cluster we wouldn't load up every core on the node because of memory contention. But hyper-transport with NUMA completely negates the need to do that. We routinely fully load the nodes on the new machine without any trouble at all.

If AMD keeps it up they are going to find a lot of business in the high-end computing segment.

Re:My Company Contributed to This (0)

Anonymous Coward | more than 3 years ago | (#36580376)

For a little balance, Intel's press release from last week:

"TOP500 Supercomputers
The 37th edition of the Top500 list, which was announced at ISC, shows that Intel continues to be a force in high-performance computing, with 387 systems or more than 77 percent, powered by Intel processors. Out of all new entries to the list in 2011, Intel powered systems accounted for close to 88 percent. More than half of these new additions are based on latest 32nm Intel Xeon 5600 series processors which now alone power more than 35% of all systems in TOP500 list, three times the amount comparing to last year."

http://newsroom.intel.com/community/intel_newsroom/blog/2011/06/20/intel-equipped-to-lead-industry-to-era-of-exascale-computing [intel.com]

Re:My Company Contributed to This (0)

Anonymous Coward | more than 3 years ago | (#36580422)

Hyper transport + NUMA isn't what allows that scaling, it's really NUMA + Directory ("HT assist" or also called "probe filter")

With the Directory, HT is barely loaded at all in a perfect NUMA workload because the machine knows it doesn't have to snoop the other sockets

Re:My Company Contributed to This (1)

The Finn (1547) | more than 3 years ago | (#36581152)

I was under the impression that intel's answer to hypertransport was QPI, introduced with Nehalem and continued in Westmere and Sandy Bridge. Intel may employ me, but I still root for AMD and wish I could buy a quad-core VIA.

Re:My Company Contributed to This (1)

gman003 (1693318) | more than 3 years ago | (#36584010)

QPI is only used on multiprocessor-capable Xeons and a very few "enthusiast" desktop processors. Sandy Bridge uses something called "DMI", which is essentially a PCIe x4 bus. Only the Nehalem i7s, the Westmere and Gulftown Xeons, and the Tukwila Itaniums use QPI. Even some of the Xeons use DMI - the smaller, single-processor-only Xeon E3s.

Interestingly, AMD has a similar bus - UMI, which is used on the new "Fusion" processors. Which are designed for use in tablets and smartphones, not computers, so HyperTransport is overkill.

News for nerds (0)

Anonymous Coward | more than 3 years ago | (#36580262)

But does it matter?

Excuse my insolence to the goddess of inconsequence, but who cares about supercomputer hardware or OSs and other such arbitrary Top500 lists?

Not to say there aren't cool stuff to talk about when it comes to supercomputers. Like for example, brilliant new ideas to parallelize things and clever ways to write distributed systems, new solutions to problems that could only be solved synchronously, interesting new applications for such systems etc etc. Whether they prefer to use Intel or AMD is not very exciting news, and tbh sounds like corporate marketing.

IMHO.

Nice to see you're still competing (1)

ErikZ (55491) | more than 3 years ago | (#36580458)

But when are those 28nm CPUs coming out?

Mostly for your ATI video cards. The heat output of your 6000 series is incredible. Love the industrial fan though.
 

opetaing sysytem (-1, Offtopic)

xhwbyyh (2312900) | more than 3 years ago | (#36581334)

I have installed Windows 95,98,ME,XP.All have been a pain in the assss.Hours of updating,and down loading this and that just to get it all to work right.With Windows 7 it loaded the first time.And with all the updates?It does it for you. http://www.buywindows7s.com/ [buywindows7s.com]

Moster beats by Dr.Dre solo hd (1)

shenzhenelectronic (2122228) | more than 3 years ago | (#36581368)

Hi ,this is shenzhenelectronic. Our website is http://www.shenzhenwholesale.com/ [shenzhenwholesale.com] we have some electronics ,for example Microsoft Xbox360 Live card 10000 US PointsMoster beats by Dr.Dre solo hdMoster beats Dr.Dre pro and so on. Welcome to visit our website Moster beats by Dr.Dre solo hd http://www.shenzhenelectronic.com/monster-beats-by-dr-dre-solo-hd-headphones-black-red-white-hot-sell_p53915.html [shenzhenelectronic.com] Moster beats Dr.Dre pro http://www.shenzhenelectronic.com/brand-new-monster-beats-by-dre-pro-headphones-in-white-black-5pcslot-hot-selling_p53932.html [shenzhenelectronic.com]

I want to like AMD. (3, Interesting)

TheRealQuestor (1750940) | more than 3 years ago | (#36581582)

I really honestly DO want to like them. And I don't really hate them any more [mostly since I stopped using them about 4 or so years ago] And to be honest for the dollar AMD does quite well against Intel for most needs. My problem with AMD systems are a much MUCH higher failure rate for the motherboards using the same brand, quality, and price point as the intel bretheren:( and they are slower.

My biggiest issue is I am VERY VERY impatient when it comes to computers and I can't find anything from AMD that even comes CLOSE to my current i7 2600K @ 5Ghz. Not even a mb/cpu(s)/memory costing 3 times as much can touch this thing in most everyday things I do such as read/program/play games/photoshop/repeat. I finally, a few weeks ago, broke down and gave them another chance and built "my" 1st amd rig in like I said about 4 or so years and whilst it's fine for the money [it's a Phenom II X4 955 Black, water cooled, overclocked to 4.2Ghz with 4 gigs DDR3 @ 1333]. it's still much slower than the i5 I built around the same time [i5 750 @ 3.4Ghz, and 12 gigs DDR3 @1328 air cooled] and both cost about the same in parts, yet the i5 SMOKES the 955 even at 800Mhz less per core. So dollar for dollar you're still "faster" with intel and I Like fast. I pay well for fast. IF Amd was faster I'd be using them right now but they don't have ANYTHING to offer me except the HD 6990 I have in here but that's still ATI in my minds eye.

AMD will need to pull a major magic rabbit out thier hats on this upcoming bulldozer [which does seem to have intel worried as they are delaying the x78 chipset and subsiquent LGA2011 cpu's and boards until AMD reviels thier hand]. Smart on thier part I guess, sucks on mine cause I couldn't wait and went with the flakey p67 but hey in a couple of months I'll have a fairly cheap motherboard/cpu combo for sale :)

i just don't see it.

Re:I want to like AMD. (0)

Anonymous Coward | more than 3 years ago | (#36581856)

Maybe it's 'slow' due to a lack of RAM? 4 gig for the AMD and 12 for the Intel?
Hardly seems fair.

Re:I want to like AMD. (0)

Anonymous Coward | more than 3 years ago | (#36581896)

It *is* fair if (as he says) the cost for both machines is the same.

Re:I want to like AMD. (1)

Rockoon (1252108) | more than 3 years ago | (#36583852)

If he was trying to build a performance AMD system, he wouldnt have purchased a Deneb-based 955. Hell, even if he built the thing a year ago he could have gotten a Thuban instead of a Deneb.

The translation is "I purchased the very latest Intel design but not the latest AMD design, and then sunk a shitload of money into a water cooling system trying to get that bad CPU choice running faster"

Hence, its not a fair comparison as the grandparent noted. What the great grandparent just demonstrated was his own extremely bad choices.

Re:I want to like AMD. (1)

TheRealQuestor (1750940) | more than 3 years ago | (#36593448)

within about 50 dollars of each other I got the 955 for like 130 and I got a deal on the i5 which ran me about 179.

And both motherboards where about 110 and I originally had 4 and 4 but added an extra 8 to run a VM. on an average system there is NO speed difference from 4 gigs to 12 gigs. NONE. I use the extra 8 gigs to run a linux vm on a OSX host. I've benched the piss out of them and whilst I did not expect the AMD to come anywhere NEAR my i7 2600k I DID expect it to run nearly as well as a i5 running nearly a Ghz slower and it didn't even come close.

Re:I want to like AMD. (1)

markass530 (870112) | more than 3 years ago | (#36582070)

your math sucks. your phenom is over twice as cheap as the i5, and you gave it 1/3rd the ram. also good AMD MB's are around 50-100+ bux cheaper than intel boards.

Re:I want to like AMD. (3, Informative)

Ecuador (740021) | more than 3 years ago | (#36582984)

Hmm, you make a couple of very bad points.
1) The "MUCH higher failure rate for motherboards" is something I have never heard before. Especially when you go on saying "at the same price point as Intel" - are you still referring to motherboards, it is very hard to find Intel motherboards as cheap as AMD. Anyway, I have experience with hundreds of systems both Intel and AMD based over the years and I can't agree there is any sort of significant failure rate on motherboards of any of the two (although I have seen many integrated ethernet controllers go bad). With one exception. Around 2003-2004 a lab bought a dozen Dell Precision slim desktops with a Prescott. When I saw them I told the people who had ordered them that I found it suicidal to put 3.2GHz P4s (I don't remember if they were Prescott or Northwood) in such small enclosures. Sure enough, about 1.5 year later half of them had blown their motherboards, I don't know how it went from there.
2) Your basic argument is that regardless of how much money you throw on AMD, you can't get the performance of Intel, which is, well, moot as AMD unfortunately (for consumers) is not competitive in the high-end. Then you go on comparing an i5 750 which is TWICE the price of the Phenom 955 (newegg: $214 vs $113 - free $15 gift card), has much more expensive motherboards and you pair it with 3 times the RAM. Yet, *surprisingly*, the i5 is faster. Gee, who would have guessed? AMD still is price-competitive, at the price points they cover there is nothing you can get from Intel that has the same performance (esp. if you include motherboard price).

Furthermore, I would like to add that for some of us that run custom 64bit software, AMD still seems to hold strong. Example: Last year I built a Phenom 955 based system which went for under $1000, even if it had the highest quality components (best mobo with USB3, eSATA, my favorite Antec case & PSU etc). I chose an AMD on an otherwise not budget system, since it would serve mainly as a HTPC. Half a year later I bought a $4000 Mac Pro with a 3.2GHz Bloomfield quad-core Xeon. Guess my surprise when most of my own software (most doing string processing in C and Perl) are about 10% faster on my AMD-based HTPC!!! The only way to get more performance out of the several times more expensive Xeon is to manage to get over 4 concurrent threads running so I can get some benefit from HT! Not to mention that for a that much money the Mac Pro isn't even giving me USB3 or esata, which is absurd... but I digress... the point is that AMD machine which was not chosen for its high-end computation power, still holds strong against some of Intel's finest, even if more common workloads show a big preference to Intel.

In the end, I hope AMD gets back in the game. I always buy what is the best (or best value) at any time, and, historically, AMD has been my choice more often than not, but the only reason we have things like Sandy Bridge coming out of Intel right now is because AMD is pushing them. In fact, if Intel's big pockets had not prevented AMD from dominating the market like they were dominating in performance during the 2000-2005 era, we would have a much stronger AMD now and more competitive products for us consumers.

Re:I want to like AMD. (1)

TheMiddleRoad (1153113) | more than 3 years ago | (#36585498)

Umm, buying an Apple machine isn't exactly making a decision based upon value, but you're making a value comparison. As for speed, sure, there are some tasks where AMD's architecture will work better, and like you said, you can optimize for either. For most users, the Intel architecture is just superior for price/performance. In some cases, AMD is better, usually when you're looking for value in highly parallel tasks, but not outright speed. Intel is loafing along, not pushing their market dominance, waiting for AMD to launch their next generation. Then you will see a big jump in Intel's speed.

Re:I want to like AMD. (1)

Ecuador (740021) | more than 3 years ago | (#36586130)

I bought an Apple machine because it was required for my work, obviously it was not based upon value. But if you read that part as a direct comparison of a Xeon to a Phenom, you did not get my point.
If you did want a price/performance comparison of the two CPUs, you would have to note that the cheaper consumer version of Xeon, the i7 @ 3.2GHz costs a bit more than 2.5X the Phenom. so you would need almost 3X the performance for it ti make a better value.
The Intel architecture currently is superior. As such, they don't have to lower its price as much as AMD does, so the AMD solutions are almost always a better value at their price points. But if you can spent a little more money...
As someone who has been buying CPUs since the 386/486 era, I will tell you that for me the greatest thing AMD has done for the consumer is that Intel is currently selling CPUs that are close to their flagship at under $300. Up until the Athlon, if you wanted an Intel CPU that was not previous gen or crippled in some severe way, you would have to pay dearly ($1000+).

Re:I want to like AMD. (2)

Kjella (173770) | more than 3 years ago | (#36583768)

AMD will need to pull a major magic rabbit out thier hats on this upcoming bulldozer [which does seem to have intel worried as they are delaying the x78 chipset and subsiquent LGA2011 cpu's and boards until AMD reviels thier hand].

I think it's been quiet well established for some time now that Intel owns the high end. That they're delaying is probably more because their $2-300 CPUs are pounding AMD pretty good already. If as usually Q3 means the very end of Q3 there's still a full quarter where Intel doesn't need to lift a finger to win.

Re:I want to like AMD. (2)

gman003 (1693318) | more than 3 years ago | (#36583812)

The way I tend to look at it is "AMD gives you the most bang for your buck; Intel gives you the most bang". AMD CPUs are cheap - I got a Black Edition (unlocked multipier, high-end) Phenom II X3 for about the cost of a low-end Core 2 Duo. It's more than fast enough for most usage, and the cost is great. But AMD can't really compete with the highest-end Intel processors - their hexa-core Phenoms are roughly a match for the quad-core Intels, and they can't beat the hexa-core i7s. Server-side, Intel isn't afraid to just throw cache at a processor - the 8-core (16-thread) Xeons max out at 24MB of cache, over double the most expensive 12-core Opterons.

If you've got money to burn, or if you justifiably need more power than AMD can give you, or even if you're just compensating for something, Intel's a good choice. But I can see why AMD is making progress in supercomputers - huge machines like that, you kind of have to assume that things scale to more processors. So AMD's processors might only provide 75% the performance of Intel's, but they do it at 50% the cost. Scientists like how that math works out.

I've also noticed that AMD is moving into businesses pretty well. A lot of simple work desktops seem to be sporting an "AMD Business Class" sticker lately.

But, AMD performed poor in BAPCo? (1)

G3ckoG33k (647276) | more than 3 years ago | (#36581770)

This Top500 comes in handy after these:

http://www.brightsideofnews.com/news/2011/6/24/amd-insiders-speak-out-bapco-exit-is-an-excuse-for-poor-bulldozer-performance.aspx [brightsideofnews.com]
Following our coverage on AMD's exit from BAPCo and blog post made by Nigel Dessau, we got a surprising call from the person at the heart of AMD which we had to check out. After the end of an eye opening conversation, we started calling our sources in order to confirm if the claims made by an obviously disappointed engineer hold any substance. We talked to our usual sources inside the company, as well as with a number of sources at their key partners and customers. The odd part was that all of our contacts said the same thing - the story checks out. Thus, we bring you the modestly edited version of our conversation, filed with comments.

AMD's BAPCo Exit is a Smokescreen
First and foremost, we started the discussion over the blog Nigel Dessau, AMD's Senior Vice President and Chief Marketing Officer wrote, stating clear reasons why AMD decided to leave the BAPCo and why AMD considers SYSmark 2012 an invalid benchmark.

"When I read Nigel's blog and saw the press release from BAPCo it made me sick because our CMO talks about transparency and honesty and it's all smoke and mirrors. At the end of the day, we actively had internal teams and external organizations hired to promote/discredit SYSmark. Not because it was inaccurate, but because it is accurate. Back in the original Athlon 64 and Opteron days, when we were winning in SYSmark we were heavily promoting it in the public sector, who in turn used it as a benchmark on which they based many of their purchases on. It was us who actually got BAPCo and SYSmark inside several government tenders to win orders measured in tens of thousands of systems. SYSmark was used to show how our K8 processors were beating Intel's NetBurst."

http://www.zdnet.com/blog/computers/why-did-amd-quit-bapco-board-poor-bulldozer-performance-on-sysmark-2012-or-intel-bias/6230 [zdnet.com]
The latest dust-up in the AMD-versus-Intel never-ending conflict concerns BAPCo, a consortium of tech companies that releases a set of benchmarks, including, most importantly, SYSmark. This week, AMD quit the BAPCo board, and speculation over why has run rampant ever since.

Officially, AMD claims that the latest version of SYSmark, the just-released SYSmark 2012, fails to keep up with current computing trends and ignores the increasing role the GPU plays in computing tasks. Since AMD is trying to differentiate itself from Intel by boosting the GPU in its new chip designs, SYSmark’s reliance on just the CPU, in AMD’s opinion, doesn’t reflect everyday computing performance.

That’s the official word. But conspiracy theorists think there’s more to the story than just that. Most sensationally, Bright Side of News has run a piece with startling claims from “unnamed sources,” most notably that AMD decided to pull out of BAPCo because its forthcoming Bulldozer chips delivered underwhelming performance on SYSmark 2012, and that the company has spent resources toward surreptitiously undermining BAPCo through negative PR campaigns. According to the piece, AMD’s paranoia about SYSmark is related to the benchmark’s role in securing government contracts and the chip company’s fear that it won’t win new contracts with poor SYSmark 2012 results.

Coincidence?

Re:But, AMD performed poor in BAPCo? (0)

Anonymous Coward | more than 3 years ago | (#36583186)

I might believe your conspiracy theory a bit more if VIA and NVIDIA didn't also pull out of BAPCo based on the same reasons AMD stated. I might also believe your conspiracy theory even more, if SYSmark2012 wasn't compiled using the Intel compiler which still to date (even after losing a massive lawsuit to AMD) cripples code ran on non-Intel chips.

Robert Anton Wilson stated that holding onto a conspiracy theory only makes sense when all other alternatives have been tried and tested, and discarded being more ludicrous than the stated conspiracy theory. Another less known person once stated (possible more colloquially), "Facts are irrelevant when they contradict opinion".

Strange (0)

Anonymous Coward | more than 3 years ago | (#36583484)

Strange that the reporters chose to miss out on both VIA's and nVidia's departures.

Food for yet another level of conspiracy theories! :)

Anandtech cites AMD, nVidia, and VIA (0)

Anonymous Coward | more than 3 years ago | (#36586806)

http://www.anandtech.com/show/4464/amd-resigns-from-bapco-over-sysmark12-concerns

"Update: AMD Resigns from BAPCo Over SYSmark 2012 Concerns; NVIDIA & VIA Also Leave, BAPCo Responds"

Re:But, AMD performed poor in BAPCo? (1)

Rockoon (1252108) | more than 3 years ago | (#36583966)

Also not mentioned is that BAPCo is code name for Intel. In fact, BAPCo was sued for constructing SysMark 2002 (which was then shipped directly from Intel's mail rooms when you purchased it from BAPCo) to misrepresent the P4's performance.
Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?