* * Boeing chooses AMD Supercluster * *

<A HREF="http://famulus.msnbc.com/famuluscom/businesswire03-14-061650.asp?sym=AMD" target="_new">http://famulus.msnbc.com/famuluscom/businesswire03-14-061650.asp?sym=AMD</A>
80 answers Last reply
More about boeing chooses supercluster
  1. Good for AMD.
    The system in the jets won't have any cooling problems, with all the air passing the jet. ;-)


    <i><b><font color=red>"2 is not equal to 3, not even for large values of 2"</font color=red></b></i>
  2. LOL! Yeah, I'm not sure I'd trust a passenger jet I was riding in to be AMD powered. But, like you said, if the CPU was mounted outside the plane it could potentially overlock to, what, maybe 3GHz??
  3. Or it could be left inside and used to heat the cabins!


    <i><b><font color=red>"2 is not equal to 3, not even for large values of 2"</font color=red></b></i>
  4. LOL!!!
  5. This is not for design of jets but, booster rockets to lift payloads.

    This very small cluster(96x) is for CFD, computational fluid dynamics.
    For the light duty AMD is ok and cheap since SMP is not required. CFD on booster rockets has been around for the past 30+ years. step into GFD (geophysical fluid dynamics) and AMD is no where to be found.

    Its just a monumental point for AMD actually scoring a cluster bid, boeing went the cheap route. nothing wrong with boeing saving a buck when computational power vs cost is a factor (since we been putting rockets in orbit for 30+ years). Amazing what sales people can sell these days.
  6. Awe, just because businesses are finally waking up to the performance benefits of an Athlon based system, in addition to the cost advantages, is no reason to get snipy. Or are you suggesting a 96 processor solution from Intel would beat it, costs be damned? I don't think even you would be that stupid, despite countless posts suggesting otherwise.

    /Athlon-1.2GHz@1370MHz(137MHz*10)/Asus_A7V133/
  7. To be honest, I don't think you know much about SMP. And by the way, AMD does support SMP.

    http://www.theregister.co.uk/content/archive/13891.html


    <i><b><font color=red>"2 is not equal to 3, not even for large values of 2"</font color=red></b></i>
  8. Number one consideration when going with a cluster is SMP (you look supprised). AMD is currently void of SMP.

    I can tell from your post that your clueless in this area, just a AMD lemming reply despite countless posts suggesting otherwise.

    That number of "96" could be total machines sold, meaning that 48 could comprise the cluster and rest are for engineering, management, and workstations.

    This bid came down to cost. not performance. if performance was the criteria, AMD would not have been a factor. that inferior cluster is far short of 700Gig per sec bandwidth I am used to installing.
  9. Lets here your 'technical' explanation of SMP. come on, enlighten me...


    <i><b><font color=red>"2 is not equal to 3, not even for large values of 2"</font color=red></b></i>
  10. why does everyone in the forum always have to "one up" everyone else? does it make you guys feel better to be able to put down someone you dont even know? most people like those bickering in this post try to act all high and mighty, arguing over the stupidest things ad finitum. please cut your comments down to a purer state and leave out those amd or intel directed barbs.

    My name is Mud.
  11. You've got to be kidding me!!! Do you honestly think Boeing, **BOEING**, would select AMD becuase they saved a few dollars? FUGGER, does Intel pay you to spew your worthless propoganda? If not, what motivates you? I can't even begin to understand why there are Intel droids like yourself who are so devoted to defending a company that doesn't even know you exist.
  12. Hey FUGGER, can CYBERIMAGE come out to play?
  13. ROFL @ Holygrenade!!!

    Ok, Einstien that post was from almost 6 months ago.

    Now stroll your gimpness over to www.pricewatch.com and do a search for 760MP all you want. WTG

    AMD is yet to bring SMP to the market.

    I suggest you do a little reading on the net about SMP. Last thing I want is a gimp like you trying to critique me on SMP.

    Try reading your link before you post it, and try to pick a product that is actually on the market next time. That usually helps your point.

    Make your next stop www.amd.com and search for SMP. wow you find like ZERO subjects on SMP (imagine that)
  14. Nikko, when investing in clusters your looking at a few hundred thousand to millions of dollars. not like the 20 bucks you saved.

    The reason this article stand out is that AMD and cluster were never used in same sentance before. Its really a joke since the extreme light duty of this CFD model requires.

    You have a problem with boeing saving a buck? you would rather boeing overkill this project.
  15. That sounds more like a madmans ranting, rather than an explanation into SMP.

    You really don't know what SMP is do you?

    To you it is just a buzz word.
    Do you even know what the acronym stands for?
    Do you know the difference between SMP and SMT?

    Why am I even asking you these questions? Of Course you don't.


    <i><b><font color=red>"2 is not equal to 3, not even for large values of 2"</font color=red></b></i>
  16. This is not the first cluster AMD has won. I recall reading an article several months ago about another AMD cluster.

    There is a pretty major flaw in your 'cheap' comment. In business, if a department needs to buy some hardware, they send a capital expenditure request to their corporate uppers. They look at the request, putter around with the numbers, and respond with how much they can actually spend. So in this case, Boeing set aside X dollars to spend on a cluster to analyze data. How would spending that money on an ellegedly inferior cluster be the more cost concious decision? If an Intel solution is better, why didn't they buy it?

    /Athlon-1.2GHz@1370MHz(137MHz*10)/Asus_A7V133/
  17. Fugger,
    Just for the record, IBM announced yesterday that it will be releasing Dual processor Athlon servers. While they may not be available yet, I think that is a pretty good indication that the product is both coming to market and is a viable, stable solution. But, I suppose IBM has just been suckered by all the AMD marketing hype and the Lemming mentality it has created. After all, AMD has a HUGE marketing presence (NOT).
  18. Actually Fugger, SMP is often not a consideration at all for clusters. It all depends on your application, but if you need lots of memory, memory bandwidth, and network bandwidth per CPU, you never want more then 1 CPU per motherboard. A CFD problem fits this desciption very well due to the very large matrix sizes involved.

    If you have smaller individual tasks, and stuff that can be efficiently threaded (as apposed to message passing), SMP is great and can significantly reduce the size & cost of a cluster. I may have to go with an Intel solution just because it can fit in one rack instead of two, despite the performance hit :-(

    And by the way, trans-sonic CFD is NOT an "extreme light duty" application by any means.

    My guess is Boeing went with AMD because they could get by with 96 CPUs instead of 120 Intel CPUs, and at a lower cost. In fact, this is exactly what Linux Networx reccomends for floating point compute bound applications. I know, because I was just discussing a 32 node cluster with them not 5 minutes ago. They promised to send some AMD vs. Intel floating point benchmarks with the quote.


    In theory, there is no difference between theory and practice.
    In practice, there is.
  19. Here's a link to another athlon cluster. It also shows the huge difference SIMD makes if you can use it.

    http://arstechnica.com/cpu/2q00/klat2/klat2-1.html

    In theory, there is no difference between theory and practice.
    In practice, there is.
  20. Cool Signature!


    <i><b><font color=red>"2 is not equal to 3, not even for large values of 2"</font color=red></b></i>
  21. Another cluster vendor I talked to today mentioned that they will be offering SMP Athlon nodes in a few months.


    In theory, there is no difference between theory and practice.
    In practice, there is.
  22. You can't answer me, can you?


    <i><b><font color=red>"2 is not equal to 3, not even for large values of 2"</font color=red></b></i><P ID="edit"><FONT SIZE=-1><EM>Edited by holygrenade on 03/14/01 08:52 PM.</EM></FONT></P>
  23. Fugger:

    You wrote: "The reason this article stand out is that AMD and cluster were never used in same sentance before. Its really a joke since the extreme light duty of this CFD model requires"
    Well, it appears they are begining to be mentioned more often. See this link http://www.tu-chemnitz.de/mbv/TechnThDyn/mpf/mpf_athlon_cluster.html
    They do not seem to think that their application requires "lightweight"duty.

    And This link:http://biz.yahoo.com/bw/010122/ca_amd_3.html
    for info on the Beowulf cluster.
  24. Explaining SMP to you is useless.

    Dig at all the angles you want, Ill tell you the same.
    I wont waste my time. no copy, no paste, no search for URL. sorry your on your own. stay clueless.

    Im not about to open myself up to your lame ass critique of my reply on SMP.

    Sojourn, bieng as we do not know what boeing budgeted for this project lets not assume. but we can look at Linux NetworX and see that they are "cheapest" cluster solution provider available (just short of in house techs building custom). so price was a factor.

    Boeing a big company going cheap route...

    One thing we know for sure, SMP was not a factor.
    If SMP were a factor, AMD would have not been quoted in this bid.
  25. some ppl can't wait 90+ days and prob a full yr to iron out the bugs.


    "Amd cpu...Gone in 2 secs flat, it truly is a fast chip!"
  26. ergeorge, please have them include Alpha numbers too =)

    Bummer we cannot get northwood numbers now too.

    Were talking gigflops here. not teraflops. still very low end of the spectrum.
  27. Yup, I've also spoken to a couple cluster vendors recently who said they too would be offering SMP Athlon nodes in the near future. Both vendors, as one would expect, already have working 760MP machines and they were both very excited about the performance. No performance benchmarks were offered in either conversation but both vendors indicated the performance would be phenomenal.

    University of Kentucky, University of Delaware, an Australian University and Boeing can't all be wrong, can they? It's fun watching the dominos fall..............

    You better get used to it, FUGGER/CYBERIMAGE.
  28. "One thing we know for sure, SMP was not a factor.
    If SMP were a factor, AMD would have not been quoted in this bid."

    As I explained before, SMP can be a bad idea for clusters. Depends on your application. and actually, <b>Intel IS the cheaper solution per CPU if SMP is useful to you</b>: half as many motherboards, cases, power supplies, racks, hard drives (if your not going netboot), etc.

    as I see it, there are three scenarios:
    -They could have used SMP, but AMD smoked Intel so bad that they went for the more expensive solution. Which may not actually be more expensive due to the number of CPUs saved, and price of the CPUs.
    -SMP wasn't useful for their app, and single CPU AMD smoked Intel and was cheaper anyway.
    -Boeing engineers are idiots. I've worked with Boeing, and would have a hard time believing this.

    "Linux NetworX and see that they are "cheapest" cluster solution provider available"

    You may remember these guys as Alta. They've been around longer then any other vendor of turnkey beowulf style computational clusters. They offer a far more comprhensive & integrated solution then any other vendor I've talked to (and I've talked to a few). They are definetely not the cheapest solution out there. And yes, I've talked to the SGI cluster guys. I was less then impressed for the price.


    In theory, there is no difference between theory and practice.
    In practice, there is.
  29. Nikko should go to work for Cleo on the psycic hotline.

    http://www.misscleo-psychic.com/

    cluster vendors? URL's?
  30. You know what I don't blame you. Sure, you cant answer a question to which you don't know the answer.


    <i><b><font color=red>"2 is not equal to 3, not even for large values of 2"</font color=red></b></i>
  31. Bingo! We have a winner. Ergeorge is right. Simetric multi Processing depends on the bus and the OS not the processor.
    Anim88tor
  32. Agreed, Alpha will completely smoke anything by AMD or Intel in floating point, if you can afford it. It's out of our price point :-( They also have 64 bit floating point if that's important to you. I'm not sure they'll be sending Alpha numbers, that conversation was pretty short :-)


    In theory, there is no difference between theory and practice.
    In practice, there is.
  33. Nikko,
    Those vendors you spoke with must just be more ill informed lemmings that can't afford a real processor and have decided to play with the equivalent of data tinker-toys instead. Oh, sorry.....I was possessed by the blue-man syndrome for a second. But seriously, can someone explain to me why it is so hard for certain individuals to admit that AMD is finally making strides with a good product that is gaining more and more market acceptance with every passing day? And, can someone also explain why it is that in the face of more and more facts to support the position that AMD's solution is at least competitive and in some cases superior, those individuals resort to name calling and personal attacks when they cannot support their position or refute the facts that have been presented?
    If someone showed me DATA that AMD's solution sucked wind or FACTS that showed the current data on the P4 was somehow wrong, I would listen. Too bad that I can't say the same for a few of the other posters here. Ok...I'm done ranting. :-)
    Mike
  34. First off girls, SMP is non existant still for AMD. deal with it. imaginary machines producing imaginary numbers doesnt cut it.

    Second, AMD is very new to clusters on the low end of the spectrum. (very low end). how you girls try to grasp ahold and think that AMD has the ultimate cluster config is beyond me. you guys are clueless why clusters are being built today. and what needs are involved in spec'ing one.

    We are designing high end paralell clusters to push upwards of 10 teraflops, leap-frogging IBM best offering today of 7 teraflops. At this level you will not see Intel or AMD processors.

    Midrange clusters are dominated by Intel/Alpha machines, AMD is left out completely.

    Low end AMD based clusters are starting to make an entrance.

    Untill the 760MP chipset is released and proven, you might as well STFU about something that doesnt exist. your predictions are without merit due to the fact that the product doesnt exist.

    Now that you girls are so into researching clusters, post some URL's please.

    SETI on a cluster is a complete waste of bandwidth. so please do not go there.

    www.linuxnetworx.com
  35. A little anectdote on SMP...

    I worked on a project about a year or so ago involving a very large single threaded simulation. We had a dual PIII 733 with 1 GB ram.

    The running sim would max out at about 300 to 350 MB ram, so there was plenty of memory to run two sims simultaneously.

    If I started a single run when I went home at night, it would be done shortly after I got in the next morning. If I started two, they'd be done by lunch.

    So, in that case we'd have been better off with 2 single processor boxes. This would be a little slower then a single run on the SMP box of course, because the second CPU can do all the OS housekeeping on the SMP box. Two single CPU boxes would have been more expensive as well.


    In theory, there is no difference between theory and practice.
    In practice, there is.
  36. Firstly, i have to confess i don't know much about SMP (other than the clothing brand!) and i in fact only found out what the acronym stood for about 2 weeks ago. Since then i have found myself needing to use it more and more..

    This Boeing argument we're all having, it sounds fair to me to say that if Boeing, the LARGEST aeronautical engineering company in the world, would choose AMD over Intel, then they're a very smart company. If you think about it, AMD started off as a small-ish company that was making chips for the business side of computers. It's only lately, since basically the K7, that they've become such a heavily influence in gaming and multimedia applications. It's a well known fact that Intel are known for their business application now because they fell behind and they aren't anywhere near as good as AMD for gaming and other CPU intensive applications... Putting an Intel in a passenger jet would really scare the hell outta me.... not just coz i wouldn't trust Intel anymore as far as i could throw one of their chips, but also because... Christ with the amount of advertising they bloody do, you'd be seeing Boeing's flying through the air, and on the sides, instead of huge airline names, you'd be seeing the Screwed up faces of BLUE MAN!!!!!!

    That's all i gotta say, Blue Man = a last desperate ploy for sales

    <b><font color=blue>Note to self: Never buy <i>anything</i> without checking with <font color=red>Tom</font color=red> first...
  37. Are boeing bigger than Lockheed Martin? LM are the US Gov't's preferred Air Defence Contractor, aren't they?


    <i><b><font color=red>"2 is not equal to 3, not even for large values of 2"</font color=red></b></i>
  38. Just for the record, Fugger, no one is representing that AMD is a)not new to clustering or b)has the ultimate cluster config or c)that their multi-processing solution has been released to the public. But, I think that some of us find it interesting that AMD continues to make progress in the market, a fact that seems to increasingly iritate you for some reason. And for the record, you might find that people would give your opinion a little more (not much more in your case, but a little more) weight if you dropped the personal insults and actually argued on the facts. Who knows, you might even learn something.
  39. There haven't been to many high-end or mid-range clusters built since AMD has become competative. I wouldn't be suprised at all to see some AMD clusters in this range over the next year.

    When you're build clusters FP/$ and network performance/$ are the primary concerns. AMD & Intel have no particular advantage over each other on the latter, but for anybody doing serious computational work, there is not much question on the former.

    As you suggest, if cost is no object and you need absolute performance, then neither AMD or Intel is a contender. You go with Alpha.


    In theory, there is no difference between theory and practice.
    In practice, there is.
  40. FUGGER you are pretty crazy. SMP is not existent for AMD consumer chips, but Athlons were designed to be able to have up to 16 processers per motherboard I believe, whereas Intel was designes to support fewer. I am sure that if a company wanted to build huge clusters or SMP systems they can and will develop their own means for setting them up, even if it is currently not available in the consumer marketplace. It still confuses my why you have such a hard on for Intel and wich AMD chips were your toilet. You're nuts man.

    "We put the <i>fun</i> back into fundamentalist dogma!"
  41. Actually, I wouldn't fly in an airplane controlled by a consumer grade CPU from Intel <b>OR</b> AMD. We're talking completely different scales of reliability here relative to what is used in aerospace.

    Second, I kinda like the blue guys. If you have to watch a commercial, those are pretty entertaining. I particularly like the one with the paint & catapult :-)


    In theory, there is no difference between theory and practice.
    In practice, there is.
  42. George,
    Since you actually seem to know what you are talking about...you mind if I ask what you do?
    Mike
  43. This is not the first cluster for boeing and its not the first CFD model.

    We know boeing can afford, SUN, SGI, or IBM if they wanted too. This is just 1 project they decided to go with AMD.

    This is like less than .1% of clusters in this world. not a major landslide for AMD.

    SGI are the masters of aerospace engineering systems (Thanks JPL NASA), and comepletely void of AMD processors. Imagine that.

    I also noticed this project omitted the need for LGFD, low gravity fluid dynamics. something that fills the need for teraflops (excludes AMD)
  44. Mostly space systems analysis & design, with some AI thrown in for flavor. A lot of satellite constellation analysis & missile defense work.


    In theory, there is no difference between theory and practice.
    In practice, there is.
  45. Ksoth when you find one, let us know. thanks
  46. Have you ever tried to make anything like an Air Traffic control simulation? I might be looking at something along those lines in the near future.


    <i><b><font color=red>"2 is not equal to 3, not even for large values of 2"</font color=red></b></i>
  47. Hmmmmm....Guess computers have NO impact on YOUR daily life! ;-)
  48. "SGI are the masters of aerospace engineering systems (Thanks JPL NASA), and comepletely void of AMD processors. Imagine that."

    A lot of that is legacy from when SGI was the only thing out there that could handle the load, particularly in graphics. A lot of the old die-hard SGI shops out there are dropping SGI like a nasty diaper (we were one) in favor of x86 solutions because they just haven't kept up at the workstation level. The rapid maturing of the Linux OS is a big factor in this as well as the phenomenal speed increases in x86 CPUs. I ported a big sim from SGI to Linux and got a 2x speed increase over a brand new, fully loaded SGI Octane. That's on a system that cost 10% of the SGI system which gives you a 20 to 1 price/performance ratio!

    That said, SGI has its place. Nothing (other then maybe a Sun) can touch their overall system data throughput, but their days as THE graphics workstations are waning.


    In theory, there is no difference between theory and practice.
    In practice, there is.
  49. "Hmmmmm....Guess computers have NO impact on YOUR daily life! ;-)"

    I only have so much time to post cuz I'm waiting for my run to finish :-)

    In theory, there is no difference between theory and practice.
    In practice, there is.
Ask a new question

Read More

CPUs AMD