Slashdot Mirror


Supermicro Announces Quad-Opteron 1U Motherboard

hpcanswers writes "Supermicro, a producer of systems for the high-performance computing market, has announced a 1U-sized quad Opteron motherboard for the OEM market. The product, which is on display at CeBIT this week, supports both HyperTransport and PCI Express. It also consumes 1000 watts of power. Supermicro's announcement is all the more interesting because the company has historically only supported Intel processors."

158 comments

  1. Uh? by Anonymous Coward · · Score: 0

    Does the board alone consume 1kW? Or did the editors screw up again?

    1. Re:Uh? by r2q2 · · Score: 0, Flamebait

      RTFA. It says it requires a 1 KW power supply.

      --
      My UID is prime is yours?
    2. Re:Uh? by symbolset · · Score: 2, Informative
      Recommended 1KW power supply is not the same thing as consumes 1KW of power. The motherboard itself almost certainly consumes less than 100. It's the other stuff you might choose to plug into it that starts burning the watts. Four of this http://www.amdcompare.com/us-en/opteron/details.as px?opn=OSA850AVWOF, for example, would max out at 356W TDP. By the time you add VR efficiency and PSU efficiency into that equation you're pegging >550W of wall power just for the processors.

      If you actually built a rack full of 22-24 1KW 1U servers you should probably apply some forethought to how you're going to get rid of more heat than is generated by four of these:http://www.paragonweb.com/TNF1613.cfm. Also, you should probably have the power company increase the multiplier on your meter, lest it spin like an AOL CD on an angle grinder: http://homepages.newnet.co.uk/martynarnold/aol.htm

      --
      Help stamp out iliturcy.
    3. Re:Uh? by canuck57 · · Score: 1

      If we are going to run 1000W, EPS 12V etc.

      Why not use a quad socket mobo that can have dual procs for 8 cores? Sun and Tyan both have them. Supermicro is on the catchup.

    4. Re:Uh? by jargoone · · Score: 2, Insightful

      RTF Title:

      Supermicro Announces Quad-Opteron 1U Motherboard

  2. Historically huh? by Anonymous Coward · · Score: 4, Informative

    Funny, I've got 20 1U and 2U supermicro opteron servers. Are you sure you researched this statement?

    1. Re:Historically huh? by the+melon · · Score: 2, Interesting

      The article at the Register correctly states that they have just kept there Opteron stuff hidden. Aparently hpcanswers didnt read the whole thing or do any research.

      There AMD64 boards and systems can be found at http://www.supermicro.com/aplus

      I personally have one of these http://www.supermicro.com/Aplus/motherboard/Optero n/nForce/H8DCE.cfm and it is a quality product.

    2. Re:Historically huh? by krbvroc1 · · Score: 1

      I too am looking at a Supermicro box right now with Dual Core Dual Opteron. However, go look at their website. Select 'Motherboards' or 'Products' or 'Systems'. There is no mention of AMD! Historically Supermicro has produced a lot of Intel systems, especially Xeon SMP--that is their bread and butter. Go look at their recent Press Releases - no mention of AMD.

      I wonder if they have a marketing agreement with Intel.

    3. Re:Historically huh? by diegocgteleline.es · · Score: 2, Informative

      Yes - but A year ago

    4. Re:Historically huh? by darkmeridian · · Score: 1

      Funny, I've got 20 1U and 2U supermicro opteron servers. Are you sure you researched this statement?

      Are you sure you read researched your statement? From TFA:

      "Supermicro continues to hide its Opteron-based gear. You won't find any mention of the boxes on Supermicro's homepage or product page. It's an all Xeon affair.To dig deep on the Opteron front, you need to head here. A-plus, ya'll."

      --
      A NYC lawyer blogs. http://www.chuangblog.com/
    5. Re:Historically huh? by PenGun · · Score: 1

      Nice board. Very similar in capability to my nForce4 but probably a bit higher grade. What processor(s) have you? I have my Opty 165 up now which was my original plan, not the 165 per se, but a dualy. You can go 4 cores ;).

          PenGun
        Do What Now ??? ... Standards and Practices !

    6. Re:Historically huh? by nacturation · · Score: 1

      The statement in question was from the summary: "... because the company has historically only supported Intel processors."

      --
      Want to improve your Karma? Instead of "Post Anonymously", try the "Post Humously" option.
  3. 1000 Watts of power!??!?! by Kittie+Rose · · Score: 2, Insightful

    Isn't that equal to 1 Kilowatt? First we have ONE TRILLION BYTES, and now ONE THOUSAND WATTS! One million shames on you, Slashdot.

    --
    EpiAdv - if you like Pokey the Penguin, try this comic!
    1. Re:1000 Watts of power!??!?! by ex-geek · · Score: 5, Funny

      A Kilowatt would be 1024 Watts. So shame on you!

    2. Re:1000 Watts of power!??!?! by j235 · · Score: 1

      A Kilowatt would be 1024 Watts
      That would be a Kibiwatt, which would just be stupid to use.

      Kilo=1000. Kibi (kilo+binary) = 1024

    3. Re:1000 Watts of power!??!?! by MK_CSGuy · · Score: 1

      One million shames on you, Slashdot.
      Well, isn't that equals to one mega shames on slashdot?

    4. Re:1000 Watts of power!??!?! by nroose · · Score: 1

      Yep. Kibi is stupid to use.

      By convention, only because computers are binary, when we refer to memory, we use kilo to mean 1024 (http://en.wikipedia.org/wiki/Kilobyte). But kilo comes from the Metric System (http://en.wikipedia.org/wiki/Kilo), and means 1000. A kilowatt is most definitely exactly 1000 watts (http://en.wikipedia.org/wiki/Kilowatt).

    5. Re:1000 Watts of power!??!?! by ePhil_One · · Score: 0
      Isn't that equal to 1 Kilowatt?

      But then you lose accuracy (significant digits). 1000 Watts implies almost exactly 1000 watts, whereas 1 kilowatt could be 1.2 killowatts, or .8 kilowats, etc. Now you could say 1.00 kilowatts, but then you run into the next issue, which is that convention uses simple watts, meaning your reader will quickly understand the facts if you phrase it as 1,000 Watts, whereas phrasing it as 1.00 kilowatts requires thr reader to translate the dimensions before they understand it. Sort of like telling you my modified segway can do 30 meters per second, how long would it take you to realize that's hiway speeds.

      Then again, maybe you just thought someone would say no, a kilowatts a totally different thing

      --
      You are in a maze of twisted little posts, all alike.
    6. Re:1000 Watts of power!??!?! by Glonoinha · · Score: 1

      You really don't want to go there.

      --
      Glonoinha the MebiByte Slayer
    7. Re:1000 Watts of power!??!?! by ex-geek · · Score: 4, Funny

      Kilo=1000

      Yeah, right. That's what the power companies want you to believe.

    8. Re:1000 Watts of power!??!?! by Firehed · · Score: 1
      Hm? I think not. 1000w is exactly the same accuracy as 1kW (501 to 1049w). 1000. watts, on the other hand, indicates it being anywhere from 999.5 to 1000.4 watts. While 1000 watts implies very near exactly one thousand watts, it's scientifically not the same as 1000. watts. That decimal point is extremely important. Perhaps 1.0x(10^3)w would be the best compromise, as it's both in scientific notation and gives a range of 950. to 1049w of power consumption.

      Of course, I'm just being anal again. We all knew what you meant. Or I'm being forgetful and would fail a quiz on sig figs if I had to take another. But I'm pretty sure that's how it worked. Luckily, we tend to not measure power consumption in such a scientific form, and with the general assumption that accuracy to the ones place is implied. Especially since that decimal point would be confused with a period in a great number of cases, thus making things unnecessarily confusing. Perhaps that's why the rest of the world uses a comma instead of a period as their decimal indicator. Maybe we need to change it to a bottom-of-the-line caret mark or something.

      --
      How are sites slashdotted when nobody reads TFAs?
    9. Re:1000 Watts of power!??!?! by Anonymous Coward · · Score: 0

      sig figs? You use error bars

      1000 +- 15W or whatever

      You can also split the error,

      1000 +15-10W

      But no one uses sigfigs anywhere bug grade 10 science.

    10. Re:1000 Watts of power!??!?! by JazzLad · · Score: 0

      I think perhaps 1000 thousand shames would be more appropriate.

      --
      "If you have nothing to hide, you have nothing to fear." - Every fascist, ever
    11. Re:1000 Watts of power!??!?! by (negative+video) · · Score: 1
      By convention, only because computers are binary, when we refer to memory, we use kilo to mean 1024 ...
      No, we don't. Much, if not most, solid-state memory these days is non-binary, and does not store numerals in power-of-two-sized chunks.
    12. Re:1000 Watts of power!??!?! by cerebis · · Score: 1
      Yes, and that was the motivation for the base 2 prefixes such as kibi. Too many manufacturers were using the metric equivalents interchangably as both base 10 and base 2. This made a mess of an otherwise entirely consistent and well conceived system of measurement.

      Even though the new units were introduced years ago, they have been taken up quite slowly. However, I believe if you look closely you'll find that some manufacturers do use the new units, and that it regresses back in incorrect usage downstream on the supply chain.

    13. Re:1000 Watts of power!??!?! by Anonymous Coward · · Score: 0

      A thousand watts ought to be enough for anybody.

      -Bill Gates (1981)

    14. Re:1000 Watts of power!??!?! by afidel · · Score: 1

      Actually due to power factor correction it would be more than one kilowatt as measured by your power company.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    15. Re:1000 Watts of power!??!?! by Anonymous Coward · · Score: 0

      Uptake has been slow, because no one can say the new units and keep a straight face.

  4. Space heater by iamplupp · · Score: 3, Informative

    "It also consumes 1000 watts of power" 1000W seems a little high... Four dual core opterons doesent need half of that! Even less for single core. The article suggest using 55W opterons.

    1. Re:Space heater by Anonymous Coward · · Score: 0

      The other 750 is probably for the cooling.

    2. Re:Space heater by netkid91 · · Score: 0, Insightful

      Have you figured in power-eating/heat-exausting SCSI drives, some high-performance ram, probably those SCSI Drives are gonna be in a RAID, but still even after all of that I agree 1K Watts is overkill.

      --
      NO~, I read Slashdot because I think it's stupid.....
    3. Re:Space heater by CrayHill · · Score: 4, Informative

      As per TFA, they use a 1000-watt power supply. It does not consume a 1000 watts of power.

    4. Re:Space heater by imsabbel · · Score: 1

      DOH.
      And if it doesnt consume 1KW, why does it need a 1KW psu? 20% or so headroom are sensible, of course, but 4 opterons dont even burn 300W in worst case situations...

      --
      HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
    5. Re:Space heater by Homology · · Score: 4, Informative

      Few power supplies much better than 80% efficiency, so with a 1KW PSU you can expect 800 W that is usable. The Opteron 870 is rated as max 95W, so four of them gives approximately 400W. A few SCSI hard disks that may use as much as 30W at startup, much memory, lots of cooling, and whatever the motherboard itself consumes, and we have easily another 100W. While a 1KW PSU seems much, it does not seem excessive.

    6. Re:Space heater by Anonymous Coward · · Score: 2, Interesting

      Probably because they just happen to be manufactured that way now. Remarkable leaps forward in switched mode semiconductor PSUs lately. Take a look at this baby Mini ITX PSU
      Just because this device says it will deliver 1KW at peak I don't expect that was a design parameter. Most interesting is that it fits into a 1U profile, I expect this is a tiny little beast.

      Problem is this: There's a growing drive here in Europe towards "green" computing. A well designed micro-cluster of nano form factor boards can beat the pants of something like this in MIPS/Watt, that's the way forward. I think CPU manufacturers are taking a wrong road into ever more power hungry devices, pretty soon energy regulation is going to tip the cart towards the "less is more" way of thinking.

    7. Re:Space heater by mihalis · · Score: 2, Interesting
      f it doesnt consume 1KW, why does it need a 1KW psu

      My guess is the designers want to ensure excellent peak current capability. Maybe when the board switches on the transient power draw troubles lesser power supplies that nominally cover the steady state power demand. Certainly if what I've read about hi-fi amps is at all representative you are better off with plenty of headroom when trying to drive speakers with a spiky signal (that is, music), and so it seems to me a server would enjoy better reliability if the power supply was more than adequate for even worst case power draw.

    8. Re:Space heater by Courageous · · Score: 2, Interesting

      For server solutions like these, typically the power supply is REDUNDANT. I.e., it's two of 'em, with a failover capability. This is very common, particularly in rackmounts.

      C//

    9. Re:Space heater by grub · · Score: 1


      A few SCSI hard disks that may use as much as 30W at startup

      True enough. To avoid the power draw of a handful of drives starting at once, most (all?) mid-to-high performance RAID systems offer staggered spinup on drives.

      --
      Trolling is a art,
    10. Re:Space heater by ivan256 · · Score: 4, Informative

      Even dual processor 1U Xeon boxes typically have dual-redundant 1000W power supplies.

      You never know what an enterprise user is going to stick in the expansion slots.

    11. Re:Space heater by wetfeetl33t · · Score: 1

      Not only that, it'll keep your coffee warm!

      --
      Register the editry.
    12. Re:Space heater by Sponge+Bath · · Score: 1
      1000W seems a little high

      Hitting the 1000W mark allows them to use cool
      new terms like '1KW'. It sounds more hip than 1000W.
      Just like chip technologies, where 130nm sounds cooler
      than .13 micron.

      And what is with this term 'bubbler'? In my day we
      called it a bong. First time I heard this I expressed dismay,
      but the young people just laughed.

      Kids today! If it wasn't holding up my onions,
      I would use my belt to smack some sense into them.

    13. Re:Space heater by rahmrh · · Score: 1

      Power consumption depends on the type of CPU's and the type of RAM. I work for an OEM and have tested one of those quad motherboards with 4 - 852 cpus, and 64GB of ram, and the power consumption was under 600 watts, power consumption varies depending on the amount and type of ram and manufacturer of the actual ram chips, and the type of cpu used. They do seem to work nicely, the 64G setup won't fit in a 1u setting, the 32G setup will fit in the 1u chassis.

    14. Re:Space heater by moosesocks · · Score: 1

      Startup load isn't an issue with SCSI hard drives. Any half-decent SCSI controller can be programmed to spin drives up one at a time. From what I understand, this is the default behavior for most RAID controllers.

      Parallel ATA does not have this capability (drives spin up the instant the power is turned on). I don't know how SATA handles this....

      (For the uninformed, hard drives require a significantly higer amount of power to initially turn on than they need to operate once the platters are spinning. A hard drive that takes 30W to spin up should only take about 15W to operate)

      --
      -- If you try to fail and succeed, which have you done? - Uli's moose
    15. Re:Space heater by Ghostx13 · · Score: 1

      [ot] Actually, a bong and a bubbler are two different things, at least in the US. A Bong is (from dictionary.com)"A water pipe that consists of a bottle or a vertical tube partially filled with liquid and a smaller tube ending in a bowl". A Bubbler on the other hand is a small, hand held pipe that has a small resivoir under the bowl. Stupid kids, don't even know what their smoking out of.

      Bong:
      http://www.drbongs.com/acatalog/glass_bong_wb-17.j pg
      Bubbler:
      http://www.drbongs.com/product_descriptions/pipes/ glass/bubbler_1.htm

    16. Re:Space heater by imsabbel · · Score: 1

      If you mean with "dual redundant 1000W" 2 redundant 500W PSUs, then i agree.
      If you mean 2 1000W psus, then you are posting bullshit.

      --
      HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
    17. Re:Space heater by imsabbel · · Score: 1

      the 1000W are on the output side, not the input one...

      --
      HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
    18. Re:Space heater by ivan256 · · Score: 1

      I mean two 1000 watt supplies, though 750's are common too. I have yet to see a name brand 1U dual Xeon server with 500's. By name brand, I mean Intel whitebox, IBM, Dell, etc... The Supermicro 1U dual Xeons use two 750's inserted from the back. The Intel whitebox servers (rebranded by tons of places) use dual 1000's, one in the front and the other in the rear, but they may offer 750's as an option.

      Just because there's a 1000 watt supply in there doesn't mean the machine is pulling 1000 watts. Without any accessories, fans on low, etc, they only pull about 190 watts.

    19. Re:Space heater by ivan256 · · Score: 1

      Oh, 190 watts is at idle, of course.

    20. Re:Space heater by evilviper · · Score: 3, Informative
      BAD MODS! NO COOKIE!

      Few power supplies much better than 80% efficiency, so with a 1KW PSU you can expect 800 W that is usable.

      No, NO, NO, NO, NO! It doesn't matter if you have a 10% effecient power supply, or a 100% effecient power supply. A 1000WATT power supply will OUTPUT 1000WATTS. The difference in effeciency is how much INPUT power it will need to do that, and how much waste heat it will produce in the process.

      Power supplies are not, and have never been, rated by their INPUT. That would be just stupid, as it would seriously penalize those companies that make more energy-effecient units.
      --
      Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
    21. Re:Space heater by jdeluise · · Score: 1

      Agreed, I built a server over a year ago at work that uses a SuperMicro case with triple redundant 760W power supply. It doesn't draw nearly that. The three power modules simply balance the load. If one fails, then the load gets balanced between the remaining two...

    22. Re:Space heater by jbplou · · Score: 1
      You do realize more than just the processors require power.
      You have
      • Fans(many in servers)
      • Hard Drives
      • RAID Controllers
      • Fiber Card(possibly)
      • NICs
      • CD/DVD ROM
      • USB Ports may be required to power something plugged in
    23. Re:Space heater by imsabbel · · Score: 1

      I just took a trip to dell.com to check out, and they dont even offer 1000w for 1U servers.
      Their dual Xeon 1U servers have 550W, with the option of dual 550W.

      Another one here
      ftp://ftp.software.ibm.com/common/ssi/rep_sp/n/XSD 00361USEN/XSD00361USEN.PDF
      IBM x server 1U, dual xeon.
      Maximum offered PSU: dual 585W.

      I think you generalisations "dual 1000W, some dual 750W, NEVER seen a brand 500W" are full of shit.

      --
      HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
    24. Re:Space heater by Eternauta3k · · Score: 0

      That doesn't mean you only get 800W, it means you get 1000W but the power supply consumes more than that (I think)

      --
      Yeah. Would you choose a neurosurgeon who pokes around people's brains in his spare time? I wouldn't.
    25. Re:Space heater by mystic688 · · Score: 1

      I have a system with Dual Xeons (Prestonia) but They are running off of 450 W. Have been doing so for 4 years.

  5. It all makes sense by Ancil · · Score: 4, Funny
    the company has historically only supported Intel processors.
    That explains it. They probably have a whole bin of 1,000-watt power supplies sitting around they need to get rid of.
    1. Re:It all makes sense by Anonymous Coward · · Score: 0

      Actually, it's just me or are opterons getting too hot as they release new models?

      AMD is releasing with more than 100W of TDP in 2007, and it's releasing a "low power" series (in june 2006?) - a sign that all opteron CPUs are as power-savvy as others.

      I don't know how intel's conroe will be playing here but it seems to me that opteron is having some dificulties to get more performance from it. Opteron has been kicking intel on the butt for a while but that's not goint to last forever, in the time amd has been getting money from the opteron intel has released prescott, now core duo and then conroe. AMD can't keep betting everything on the opteron forever, if they don't release something else intel will be ahead of amd again. From the roadmaps, it indeed looks like intel is getting the lead this year...

    2. Re:It all makes sense by urlgrey · · Score: 1
      The back story on this....
      "Tom."

      "Yah, George?"

      "I just got our quarterty inventory report--we've got TEN extra pallets of 1,000-watt power supplies! I'm steamin' mad about this--I want them priced to MOVE!!"

      "Well... uh... I don't think there's much dema.."

      "I don't give a rats tail about what you think. Do whatever it takes to move those power supplies!!"
      In news today, Supermicro announces a quad Operton motherboard for 1U chassis...."
      --
      Running 'Nix is like owning a Lightsaber. It's "a more elegant weapon for a more civilized time."
  6. Obviously .... by Anonymous Coward · · Score: 0

    Obviously the 1000 watts includes the required juice to power the attachable coffee maker.

    1. Re:Obviously .... by Anonymous Coward · · Score: 0

      Funny you should mention that. It looks like someone already designed a case for this board.

  7. Intel's dominance at play here by hirschma · · Score: 4, Informative

    Supermicro has offered AMD solutions for a quite while now - just not under their "main" brand name. If you don't know that their Aplus products exist, you won't find them. Although I'm sure no one would go on record, I'd wager that Intel has pressured a heavily Intel-dependent vendor to not promote AMD's product.

    In fact, go to SuperMicro's home page, and you'll notice no mention or links to their AMD based products.

    This isn't the first time that this has happened. When AMD first shipped the Athlon, very few board makers dared to ship Athlon solutions for fear of Intel shorting them on chipsets. I recall, but cannot substantiate, that Asus and Abit first shipped Athlon boards under a "shadow brand", much as Supermicro is doing here.

    I, for one, cannot wait to buy some of the Supermicro^h^h^h^h^h^h^h^h^h^h, um, Aplus gear.

    1. Re:Intel's dominance at play here by Robber+Baron · · Score: 1

      I've got a pair of Supermicro dual Opteron boxes with 8 drive SATA backplanes and they're pretty nice.

      --

      You're using her as bait, Master!

    2. Re:Intel's dominance at play here by jpmkm · · Score: 1

      You must have done some top-notch sleuthing to dig up that link. It was hidden pretty well in the first few words of the grandparent post.

    3. Re:Intel's dominance at play here by hirschma · · Score: 1

      You might want to wash the sand out of your crack.

    4. Re:Intel's dominance at play here by CFrankBernard · · Score: 1

      If you go to http://www.supermicro.com/ and enter opteron in the Search box, there appears to be several Opteron non-Aplus servers:

      2 SUPERMICRO, INC. - PRODUCTS | CHASSIS | 1U | SC813T+-500
      3 SUPERMICRO, INC. - PRODUCTS | CHASSIS | 1U | SC813S+-500
      4 SUPERMICRO, INC. - PRODUCTS | CHASSIS | 1U | SC813I+-500
      5 SUPERMICRO, INC. - PRODUCTS | CHASSIS | 1U | SC812S-420C

  8. FINALLY. by blair1q · · Score: 4, Funny

    Finally.

    A pizza box that will actually cook your pizza.

  9. Boring by Anonymous Coward · · Score: 0

    Wake me up when we can have 4 x dual core opterons with full virtualization support in 2U's for the price equivilent of 4 dual core consumer desktops.

    1. Re:Boring by kennedy · · Score: 1

      Well, the SunFire v40z is a quad opteron... i'm sure there will be a refresh soon enough with dual core support (if it's not there already...)

    2. Re:Boring by ZESTA · · Score: 1

      The v40z is 3U. It is actually manufactured by Newisys, which was the first AMD validated server platform. My company makes servers based on the same thing. The newer models do support the dual-core chips. Quad dual-core opterons are a sight to see in action.

      The supermicro 1U quad is definitely cool, but only for some applications. For example, it can only take 3 hard drives, and only has 1 expansion slot.

      -Randy

  10. Re:Imagine what a... by tomhudson · · Score: 3, Funny

    ..beowulf cluster of these would do.

    nothing - after they blow the breakers. 1 kw is a lot of juice. per 1U.

    So quick - convince your boss he needs one of those for his desktop, and you'll "inherit" the dual cpu you conned him into buying a few months ago so you could "inherit" his last desktop.

  11. Re:Boring - NOT! by Overzeetop · · Score: 4, Funny

    Oh, come on. You know your nipples got hard when you read the headline. 8 cores, 64GB of memory, onboard scsi with built in raid. Of course we can't afford it. You're not going to bed Keira Knightley either, but that doesn't mean she's "boring".

    --
    Is it just my observation, or are there way too many stupid people in the world?
  12. Here's the punchline, you make the joke... by Anonymous Coward · · Score: 0

    "One point twenty-one gigawatts!"

    What? Someone had to say it.

    1. Re:Here's the punchline, you make the joke... by RedACE7500 · · Score: 1

      Nice. Took me a moment to get it.

  13. Re:Boring - NOT! by l3v1 · · Score: 4, Funny

    :D Yeah, one really has to be IT-related so as not to cough up yesterday's food when reading such comparison involving Keira Knightley and a quadX2 opteron board :D

    --
    I am putting myself to the fullest possible use, which is all I can think that any conscious entity can ever hope to do.
  14. Re:Imagine what a... by l3v1 · · Score: 1

    1 kw is a lot of juice. per 1U.

    Even four of those 8xx series cpus don't consume that much, and they don't say that either, they just say it has such a psu. I don't think that even simultaenously having and using the 4 cpus, the 64 gigs of ram, the 2 scsi discs, the 4 sata discs, the 2xgig network ports and some cooling fans would consume that much.

    --
    I am putting myself to the fullest possible use, which is all I can think that any conscious entity can ever hope to do.
  15. Cluster Grid by gridengine · · Score: 0, Offtopic

    Opteron boxes are very popular in compute farms/clusters, and Sun will release the 8-socket machine soon.

  16. I liked SuperMicro... by (H)elix1 · · Score: 3, Interesting

    I've still got some SBU and DBU mainboards still chugging away. The problem was I made the switch to AMD and SuperMicro waited years to make the transition. To bad because I liked their kit and in the PII/III era, and Asus, Abit, and SuperMicro were solid recommendations to friends and family. They sort of fell off the recommendation list because of the preference of the AMD CPU's. Now that they are working back into the workstation/server market, I'll definitely take a look the next time I do a system update.

    Yes, they had a white box label that did AMD stuff. Whatever. I'm glad to see the 'pro' brand get with the program.

  17. You're a moron by Dr.+Spork · · Score: 2, Informative
    If you read the article, you'd find that each of the Opterons in it will use 55 watts. Since you're stupid, I won't take any chances and do the calculation for you: 55 X 4 = 220 Watts of power (plus a bit extra for running the chipset and drives). That's not bad for what is essentially 8 cpu's! The recommended (read the spec sheet) 1000 watt power supply is there to insure perfectly stable voltages. It would be overkill for anything except applications where maximum uptime is essential (like web serving).

    For the sake of comparison, I run a 530 Watt PSU on this system, which draws about 100 Watts from the wall. Yaay for cool-running AMDs!

    1. Re:You're a moron by owlstead · · Score: 1

      For the sake of comparison, I run a 53 Watt PSU on this system, which draws about 12 Watts from the wall. Yaay * 8 for cool-running VIAs!

    2. Re:You're a moron by wetfeetl33t · · Score: 1

      "It would be overkill for anything except applications where maximum uptime is essential (like web serving)."

      That is the idea behind such a motherboard, ie it will be used for applications where uptime is essential (after all, it's a server board). I really can't imagine someone buying a mobo like this just to write emails and surf the web. If you have enough money to buy a system with one of those mobos, you have enough money to get a big PSU.

      --
      Register the editry.
    3. Re:You're a moron by tomstdenis · · Score: 1

      Ah but that's the trick, what takes me 2 minutes on an AMD64 takes you 17 on a VIA box. And I know as I run a 1.2Ghz C3.

      VIAs are good for only a few things

      1. Text processing [e.g. email, web, usenet]
      2. Low power NAT/dhcp/dns server for a house [which is what I use mine for]

      I have Gentoo Linux installed on mine and my AMDX2 while taking 9 times the power [or TDP] can usually finish emerges in a very small fraction of the time the C3 takes.

      This is like the Crusoe vs. Rest debate. Their CPUs would run with little power but then it simply took longer to accomplish work. So while you may get an extra hour of run time you lose it to waiting for things to finish. Specially as a LaTeX using software developer :-)

      Tom

      --
      Someday, I'll have a real sig.
    4. Re:You're a moron by Anonymous Coward · · Score: 0

      Why would you run a 530W PSU in a system that obviously doesn't need it? If I'm not mistaken, power supplies seem to run way more efficient near the 80% load level and often are more efficient at full load than they are at 20% load.

      I'd find some graphs, but I'm too lazy.

    5. Re:You're a moron by Nutria · · Score: 1

      it will be used for applications where uptime is essential (after all, it's a server board).

      If uptime is essential, then you'll be using a Tier-1 8U cage with hot-swappable RAM CPUs. SPARC, POWERx or some such, running Solaris, AIX, Tandem, etc.

      --
      "I don't know, therefore Aliens" Wafflebox1
    6. Re:You're a moron by Lord+Kestrel · · Score: 1

      You should setup a cross-compiler on the AMD box, and build binary packages for your C3. Then you can just scp them over, and not spend an extemely long time building software on the C3.

      It takes a little bit of time to get setup, but it quickly pays for itself. I do the same thing here, I have a SH3 box that I build packages for on this dual Xeon. If I tried to build them on that box, it would probably take weeks to build the system.

      emerge crossdev, and read the docs :)

    7. Re:You're a moron by tomstdenis · · Score: 1

      or just use distcc and put -m32 in your CFLAGS. :-)

      Point is I don't really have a lot installed so doing a weekly [re: monthly] update isn't a significant problem.

      Mostly my point was even though the C3 lacks ALU horsepower it's more than enough to act as a highly flexible NAT/router/etc box. Certainly more flexible than say a $100 linksys router and only costs slightly more anyways.

      Tom

      --
      Someday, I'll have a real sig.
  18. how about 16 cores? by Ankou · · Score: 3, Informative

    Maby you cant fit it in 1u, but Tyan makes the K8QW that supports 8 Opterons with the M4881 add on processor board. Meaning you got 16 cores of pure powa. Go ahead, compile the internet. I'd be interested in knowing if there was anything higher than that.

    1. Re:how about 16 cores? by aluser · · Score: 1
      ibm sells this with 64 cores: http://www-03.ibm.com/servers/eserver/pseries/hard ware/highend/595.html

      or maybe you meant just opterons

    2. Re:how about 16 cores? by Anonymous Coward · · Score: 0

      Nope, I'm pretty sure 8 Opterons would be more than 1u :-P

    3. Re:how about 16 cores? by Randseed · · Score: 1

      Damn. Those boards can compile _US._

    4. Re:how about 16 cores? by Prof.Phreak · · Score: 1

      We use that Tyan board at wr0k with 8 Opteron 875s (16 cores). It's really a wonderful board/setup.

      --

      "If anything can go wrong, it will." - Murphy

    5. Re:how about 16 cores? by Bert64 · · Score: 1

      SGI or CRAY will sell you a much larger system if you have the money... 4-figure processor numbers are not uncommon.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
  19. Manufacturer says ... (.Score +2 Flamebait) by tomhudson · · Score: 1, Informative
    The specs from supermicro's site say it requires a 1kw minimum psu for stability.

    mportant Chassis Notes

    To ensure system stability, a 1000W (minimum) ATX power supply [8-pin (+12V), 8-pin (+12V) and 24-pin are required]

    That's a LOT of juice!

    1. Re:Manufacturer says ... (.Score +2 Flamebait) by PhaxMohdem · · Score: 1

      1KW Power Supply Recommended, thats not to say the system is going to be constantly sucking 1000W of power 24/7 from your outlet. It's usage will vary as loads vary, and system comfigurations vary (more/less HDD's/RAM).

      Consider this:
      1x 1000W PSU (800W realistic max output) powering 8 Opteron cores in one box and 1U of rackspace
      8x 300W PSU's powering 8 Opteron cores in 8 boxes and 8U of rackspace

      Which one would you chose? :)

      --

      The Property of One's : "The Oneitude is directly proportional to the Colditude of the one." - S.B.

    2. Re:Manufacturer says ... (.Score +2 Flamebait) by tomhudson · · Score: 1

      My point was that is a LOT of juice. Its in a 1U unit. That's obviously going to need some seriously noisy cooling.

      A lot of places, the problem now isn't space - its cooling.

      I'd still take one if they gave it to me ...

    3. Re:Manufacturer says ... (.Score +2 Flamebait) by alienw · · Score: 1

      Not really, considering that in China, a 1kW rating for a power supply apparently means it can supply 400W of actual power. Seriously. Look at the tests Tom's Hardware did on some power supplies. Few of the 500W units could do more than 350W without blowing up. You probably don't want to put more than 200W of load on the things if you want the outputs to be stable.

      Just for reference, a 2.8GHz Pentium 4 with 1GB of RAM and a single 7200 RPM hard drive draws a maximum of about 120W at full load. With a really power-hungry videocard, that might go up to 200W. However, one has to use a power supply with a rating of at least 400W. Why? Because the cheap PC power supplies are poorly engineered pieces of junk with seriously inflated ratings.

    4. Re:Manufacturer says ... (.Score +2 Flamebait) by tomstdenis · · Score: 1

      Antec is your friend. I have had bad luck with noname brand PSUs before usually in the form of them blowing up my motherboard [like literally traces would burn up].

      Antec PSUs used to cost more than others but not any more. They have strict tolerances to the specified voltages and they last a long time. An Antec Sonata II case for instance costs around 125$ and gets you a mid-tower case with a 80mm case fan and a [iirc] 400W or so power supply.

      I use Sonata cases for all my boxes and they have high uptimes even through the brownouts that frequent Mississauga [I have them on UPSes].

      Tom

      --
      Someday, I'll have a real sig.
  20. Is this Google's new brain? by Dr.+Spork · · Score: 2, Interesting
    A Supermicro source rushed to claim some of the glory tied to rumors that Google has become a large Opteron shop.

    "There's some truth to the rumors" about Supermicro supplying gear to Google, the source said. "It was happy days around here."

    This is pretty slick hardware, and given Google's recent and complete switch to AMD, this seems like a good match. Opterons are just awesome, even I'm seriously thinking of buying a 165 (slobber...).

    On an unrelated note, did you notice the chipset is made by NVidia? Wow, they've come a long way! I'm impressed to see this kind of iron from a company that used to live off pimply gamers. With PCI-X supported, I can't help but wonder about what framerates one could get in a properly multithreaded game.

    1. Re:Is this Google's new brain? by Mkaram · · Score: 1

      PCI-X and PCI-Express are not the same thing ;-)

    2. Re:Is this Google's new brain? by Dr.+Spork · · Score: 1

      Aah, oops, my bad. Thank you for the correction. I meant PCI-Express, though I noticed that the motherboard only supports 8X while most graphics cards want 16. And strangely, the intergrated video is ATi even though the chipset is NVidia. This thing is really some sort of chimera!

    3. Re:Is this Google's new brain? by LurkerXXX · · Score: 1

      Most graphics cards want 16, unless they are in SLI mode, then they need 8x each. Interestingly this motherboard has two 8x PCI-Express slots (as well as 2 PCI-X slots), but still has the ATI onboard video. If only the two 8x slots were capable of SLI, this thing could be the ultimate motherboard for very rich gamers. With a multithreaded game and 8 cores, and 32 GB of RAM, the CPU and memory sure won't be the bottleneck in the system.

    4. Re:Is this Google's new brain? by Anonymous Coward · · Score: 1, Interesting

      On an unrelated note, did you notice the chipset is made by NVidia? Wow, they've come a long way! I'm impressed to see this kind of iron from a company that used to live off pimply gamers.

      I'm impressed that supermicro is releasing a motherboard with a chipset whose main market share are game kids and who started to produce chipsets yesterday. I'm shocked thy didn't included the best option when it comes to AMD - the AMD chipsets

      The reason why Intel keeps top 1 on the server market is because there's a REAL ecosystem around them, I'm writing this from a dual cpu supermicro motherboard which features a rock-solid serverworks chipset. In Intel, you can also choose IBM chipsets (they make a chipset which makes xeons beat opterons in several benchmarks with 4 cpus or so), or a intel chipset. Since AMD hasn't been relevant except for the last two or three years, the best enterprise-oriented chipset i'd choose it'd be the amd chipset (it's what HP or SUN boxes use). I guess that not losing market share is more important than quality some times - i'll be dead before using a gamer chipset for serious usage, I'd rather buy a HP machine instead of that thing.

    5. Re:Is this Google's new brain? by ScriptedReplay · · Score: 4, Informative

      i'll be dead before using a gamer chipset for serious usage

      what you don't realise is that it's not your regular NForce4 gamer's chipset. nVidia has a separate professional line, see here to which this one (nForce pro 2200) belongs.

    6. Re:Is this Google's new brain? by MojoStan · · Score: 1
      I noticed that the motherboard only supports 8X while most graphics cards want 16. And strangely, the intergrated video is ATi even though the chipset is NVidia.
      This is a server motherboard, not a workstation board. On a server board, PCI Express x8 slots (4GB/sec) are meant for things like dual port 4X Infiniband cards, dual channel Ultra320 SCSI cards, and 2GB/sec Fibre Channel cards. On a server board, PCIe x8 slots are not meant for graphics cards because most servers only need very basic integrated GPUs.

      On recent desktop/workstation motherboards, PCIe x8 interfaces (using x16 slots) are usually a "hack" to get two PCIe x16 graphics cards working in SLI or Crossfire mode. Early dual-graphics chipsets didn't have enough PCIe lanes (around 20 per chip) to allocate for two true x16 interfaces. The newest ones do.

      If you are looking at the Opteron 1xx series (for single CPU workstations), then you should know that they are Socket 939 (like Athlon64) and not Socket 940 like the other Opterons (dual CPU and up).

      AMD Opteron Model Number Comparison
      --
      TO START
      PRESS ANY KEY

      Where's the 'ANY' key? I see Esk, Kitarl, and Pig-Up...

  21. Much better use of space by mpcooke3 · · Score: 3, Funny

    This will allow for much better use of space in data center racks.

    For example I could take out 8x 1U Intel based Dell web servers and replace it with 1 8way supermicro Opteron machine.

    The Opteron server couple probably serve content faster than the 8 Intel based Dells plus with the additional 7u space I'll easily have enough space to put a saucepan on top to boil water for tea, thereby saving myself the 80p I normally spend in the data center vending machine.

    1. Re:Much better use of space by Antique+Geekmeister · · Score: 1

      No, swapping 8 single units for one octal unit is usually a bad move for webservers. Individual operating systems can typically withstand far fewer than 1000 simultaneous connections, and when one component fails on a big, expensive, octal server, the whole thing is down and you lose 8 servers. Also, the amount of disk you can strap into an 8-way system is limited by the controllers, not the CPU speed, so for storage of actual web content, it can be very expensive or very inefficient unless you buy a great big external array.

      But the time you've totaled up all the money and necessity for failover, network resources, disk, etc., you may find that 8 cheap 1U systems is actually a lot cheaper than one very expensive overpowered server.

    2. Re:Much better use of space by mpcooke3 · · Score: 1

      I was actually being sarcastic but since you seem to have taken me seriously...

      It all depends on your web serving requirements, if you are big then the 8ways are better. You probably wouldn't actually remove older machines, but starting to use 8 way machines in a 1U slot does actually make sense if you are doing mass web serving. You want to cram in as much as possible since you are then maintaining less machines and have more cost effective use of space + more req/sec peak load handling per $$$. (of course this particular motherboard might overheat your datacenter or blow your APCs but i'm talking generally... )

      If you are doing mass web serving and are scaling laterally you have everything in RAM you wouldn't be doing much disk IO on a web server, which is for a later tier or for asyncronous processing/updating (syncronous disk IO cripples web servers). a cheap dell server(750) serving from ram can manage over a thousand requests a second doing minimal processing or using a reverse proxy cache in Java. A dual core opteron I tested on friday can do well over 5000Req/sec in a similar situation i'm sure this can be made faster in C, but i'd rather spend more on servers than learn C ;)

      Web serving pretty much scales with CPU speed once you've fixed anything that could cause disk IO (if you think the you have too much data to fit in ram, then maybe aren't spreading the data over the machines in the best way).

      Other than the Mhz rating the onchip cache size also makes a massive difference to performance.

  22. Off topic but... Why? by nroose · · Score: 5, Interesting

    Why does every server have it's own transformer/converter/rectifier/power supply (I think you know what I mean, that thing that converts from AC to DC)? It seems to me that it would make more sense for a room of servers to have a DC supply for all the computers.

    What's more, most server rooms have a power backup unit, which converts from AC to DC and back to AC again, just so that the computers can convert it back to DC. This is terribly inefficient just in terms of electricity, and it also creates a whole lot of heat, just so that we can air condition these rooms with huge air conditioners!

    It would just seem to make sense to me that the world of computing would come up with a standard for using DC, and then companies would build big power supplies that would offer redundancy, power backup, and current conditioning. It would save money, power and space.

    1. Re:Off topic but... Why? by TheRaven64 · · Score: 2, Informative

      For more information on this topic, see here.

      --
      I am TheRaven on Soylent News
    2. Re:Off topic but... Why? by Anonymous Coward · · Score: 0

      Maybe because if one power supply dies, I have a server down. If your mass transformer falls over, my whole damn datacenter gone.

    3. Re:Off topic but... Why? by Rick+Genter · · Score: 3, Informative
      It would just seem to make sense to me that the world of computing would come up with a standard for using DC, and then companies would build big power supplies that would offer redundancy, power backup, and current conditioning. It would save money, power and space.


      In fact, in the telco world, this is exactly how it works. The standard is to use -48vDC. Sun (among other manufacturers) makes servers that run directly off of DC (the Netra 120 on the referenced page).
      --
      Don't underestimate the power of The Source
    4. Re:Off topic but... Why? by aaarrrgggh · · Score: 1

      AC servers cater to the lowest-common denominator, the guy that wants to plug it in on his desk.

      At 40kW though (really anything over 25kW in the US), a single/redundant -48V converter for the rack makes a lot of sense : you need a bunch of small 30A feeds like the Sun 20k's (12 per machine!).

      The trouble with 48V in high-power applications is that you need a lot of copper. A typical DC powerplant in the telecom world is good for a maximum of 400kW. That power plant is about 3,000 square feet (granted it is an 8 hour battery), and would only feed 10 cabinets full of these boards!

      Cooling these things would be a nightmare! The best solutions today can get up to 30kW with a lot of effort - things like water-cooled cabinets. Three of these cabinets would need their own 30-Ton CRAC unit which is the size of four cabinets!

    5. Re:Off topic but... Why? by camperslo · · Score: 1

      Why does every server have it's own transformer/converter/rectifier/power supply (I think you know what I mean, that thing that converts from AC to DC)? It seems to me that it would make more sense for a room of servers to have a DC supply for all the computers.

      If the voltage weren't so low and the currents so high, this might be practical, but with the large currents typically flowing on a supply of 5 Volts or less, it wouldn't take much resistance in the cables and connectors to drop the voltage excessively. Since the drop fluctuates with load, additional noise would be introduced on the supply lines also. When only one load is located some distance from the power supply, remote-sensing can greatly improve voltage regulation. That involves shifting the point that the power supply uses for the feedback signal out to the load (using a separate conductor for the signal). Since the ground return path also has a voltage drop, sensing on that side is needed too. If all loads were the same and had equal length cables, sensing feedback on just one would work. There'd still be the noise problem though. The power supply would respond to a spike in load current with a spike in voltage on the source end of the cable. The load with the current spike and feedback would see stabilized voltage, but all other loads would see an (upward) voltage spike.

      It's really annoying to see the power figure hyped. Obviously someone thinks 1000 Watts sounds impressive.
      The figure has the opposite effect on me, as my $1/month per 10 Watts continuous rule of thumb translates that to $100 on the electric bill. If one figures that a 1500 Watt space-heater runs 1/2 the time, a server that really uses 1000 Watts would be generating MORE heat. Put 50 in a rack, and have 10 racks in a room, and we're talking 500,000 Watts. Try and cool that!

      Just remember, the next time you Slashdot a site, you've melted another chunk of arctic ice.

      Maybe it's time that we phase out traditional servers and move towards some method of serving from unused capacity on user machines.

    6. Re:Off topic but... Why? by evilviper · · Score: 1
      What's more, most server rooms have a power backup unit, which converts from AC to DC and back to AC again, just so that the computers can convert it back to DC. This is terribly inefficient just in terms of electricity, and it also creates a whole lot of heat, just so that we can air condition these rooms with huge air conditioners!

      That's just completely, totally, untrue. Converting from AC to DC and back causes nominal losses.

      It's the conversion from 120v to 12v, then up to 120v again that is the real problem, and that could be easily solved if a company like APC would just create a product that uses higher-voltage banks of batteries.

      In short, it's been discussed again and again, and DC barely has any advantage at all. With a few switching power supplies getting to be 80%+ effecient (http://www.80plus.org/), their losses (including the UPSes converters/inverters) are lower than the line losses you'd get with high-amp DC distribution, not to mention how much more dangerous high-amp DC is to work with.

      There are other practical problems with DC as well. Go lift the hood of your car, and take a look at the battery terminals...!
      --
      Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
  23. It's all antique redux until PCI-X 2.0 by Anonymous Coward · · Score: 0

    It's all antique redux until they put out a board with PCI-X 2.0 @ 266mhz.

  24. Standby spare by Spazmania · · Score: 2, Interesting

    Also included: a standby spare so that when that much heat in that small a package burns itself out in six months you don't have to wait for a replacement. Supermicro is generally pretty good but packing a kilowatt of consumption into a 1U package is about as smart as running your home PC in the oven set on bake.

    Maybe I'm just sore because I've spent the last few weeks identifying the bad ram in last year's opteron rackmounts from Penguin. 2 gig ECC dimms and I'm seeing a 40% failure rate from multiple manufacturers. They stacked the damn chips one on top of another. There's no where for the heat to go. Of course they're going to fail.

    --
    Moderating "-1, Disagree" is simple censorship. Have the guts to post your opinion.
    1. Re:Standby spare by Anonymous Coward · · Score: 0

      That's typical Penguin manufacture: they get the details badly wrong in a way that you won't notice until you actually open it up and look around. I know only one manufacturer who generally makes worse hardware, and I resigned from them rather than watch them use randomly colored wires and swear they'd "tested for 3 days" hardware where one third failed the first day, they one-third of the resulting repairs failed the second day, and the last few repaired ones were only booted for 5 mintes and shipped. I saw that happen twice in a row, and resigned ASAP.

    2. Re:Standby spare by evilviper · · Score: 2, Interesting
      Supermicro is generally pretty good but packing a kilowatt of consumption into a 1U package is about as smart as running your home PC in the oven set on bake.

      A kilowatt in a 1U package is generally called a "Deluxe George Foreman Grill".

      If you pack the front and rear panels with several high-speed ear-piercing fans, you could disperse that kind of heat. You'll need to wear hearing protection when you go into your server room, but it's certainly possible.

      Still, these are the kinds of absolutely ridiculous solutions you get when co-hosting companies all decide to charge an arm and a leg because of "size" rather than making "wattage" the most important factor.
      --
      Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
    3. Re:Standby spare by IvyKing · · Score: 1
      2 gig ECC dimms and I'm seeing a 40% failure rate from multiple manufacturers. They stacked the damn chips one on top of another. There's no where for the heat to go. Of course they're going to fail.

      Cpnversely, one of the things that impressed me about the physical design of Sun's SunBlade-1500 was a shroud over the DIMM's to promote cooling of said modules - the other end of the shroud went to the chassis exhaust fan. Weird that we're having to worry about thermal management in memory - though the the intel 2167's on the Seattle Computer's first 64K board were getting a bit toasty.
  25. That's a lot of HP by symbolset · · Score: 2, Interesting
    And I don't mean Compaq.

    1 horsepower ~= 746W.

    The horsepower of our computers has gone from figurative to literal.

    One rack of these could theoretically consume more power than this: http://www.motortrend.com/roadtests/coupe/112_0107 _2000_mini_cooper_sport/

    Will noone think of the salmon? http://riversideca.apogee.net/foe/fgphe.asp

    --
    Help stamp out iliturcy.
  26. Recommed PSU rating by puhuri · · Score: 2, Interesting

    Supermicro states quite high PSU requirements. I have few 1U systems that have only 250 W systems even if the motherboard spec says minimum 350 W. However, no stablity problems even with add-on cards.

    I guess they only want to play safe and do not want anyone to complain about instability because of too weak PSU.

    1. Re:Recommed PSU rating by Anonymous Coward · · Score: 1, Insightful

      Hi, I also think that it's often possible to use lower rated PSUs, especially those of higher quality. However, smaller PSUs are more stressed since they use smaller sized components, and may not last as long as bigger rated ones.

      You can often tell a good PSU by its size, and especially weight.

  27. "1U?" by Guppy06 · · Score: 0, Troll

    What's a U? Does that mean single user?

    If it were lower-case it'd mean "unified atomic mass unit." Are we talking about picotechnology now?

    The "correct" symbol for microns is a lower-case mu. And it's more proper to use the term "micrometers." But if Slashcode can't handle ancient Greek (so much for "News for Nerds" if you can't talk math), there's always scientific or engineering notation.

    1. Re:"1U?" by tomstdenis · · Score: 1

      That's the dumbest post ever. 1U means a rackmount unit of height. Boxes are in multiples of said unit [look it up on wikipedia]. 1U is the smallest you can get [roughly 2 inches high or so]. 2U or 3U is more typical for 4P boxes, so a well ventilated functional 1U 4P box would be impressive.

      Tom

      --
      Someday, I'll have a real sig.
    2. Re:"1U?" by Anonymous Coward · · Score: 0

      That's not necessarily a dumb post. Everyone who knows what 1U means didn't know at one time. Plus it's very difficult to google for what exactly U means. Go ahead, try it

    3. Re:"1U?" by Rick+Genter · · Score: 3, Informative

      If, on the other hand, you use Google to search for 1u, the very first link gives a nice, detailed explanation on exactly what 1U means.

      --
      Don't underestimate the power of The Source
    4. Re:"1U?" by Anonymous Coward · · Score: 0

      It's dumb because it doesn't say "Gee, what's a U?" Instead it's a rant about Slashdot's supposed inability to render mathematical symbols.

    5. Re:"1U?" by Antique+Geekmeister · · Score: 2, Informative

      1U is exactly 1.75" high. It's an old standard for putting the holes in 19" and 24" racks, designed to allow devices of different heights to be stacked efficiently by having the screw holes laid out according to certain well-defined patterns. The patterns are quite old, and caused a lot of problems when the US tried to switch to metric (and basically failed in the computer manufacturing industry).

    6. Re:"1U?" by Xandu · · Score: 1

      If, on the other hand, you use Google to search for 1u, the very first link gives a nice, detailed explanation on exactly what 1U means.

      Are you feelin' lucky? Well, are you? Punk!

      --


      --Xandu
  28. "Consumes" 1000 watts "of power" by Ancient_Hacker · · Score: 2, Interesting
    Hate to be pedantic, but this pizza-box doesnt "consume" 1000 watts "of power". In many ways:

    • The sticker on the power supply may say "1000 watts", but that's an upper limit of what the power supply can do when maxed out. I suspect in typical setups the draw will be much lower. Opterons only draw about 80 watts, so a quad of them will only be about 300 watts. You're only going to max out with a lot of memory and many 10K RPM disks.
    • The unit doesnt "consume" the power, it gets mostly (99%) gets converted into heat.
    • Saying "1000 watts" is enough. No need to say "of power", the only meaning of watts is as a measure of power. I suppose we should be glad they didnt use one of the many wrong units, like watt-hours, or watts per hour.
    • That's a lot of watts in a 1U height! Do they suggest mounting it high on the rack so you can use it as a hair dryer?
    1. Re:"Consumes" 1000 watts "of power" by Aluvus · · Score: 0

      "The unit doesnt "consume" the power, it gets mostly (99%) gets converted into heat." What the hell do you think "consume" means?

      --
      Never mistake "can" for "should".
    2. Re:"Consumes" 1000 watts "of power" by Ancient_Hacker · · Score: 1

      To be precise, it consumes the electricity, but not the power. Power can be conveyed, transduced, transformed, dissipated, or converted, but never consumed.

  29. Re:Boring - NOT! by Sique · · Score: 1

    I'll take the quadX2 anyday :) Much less maintenance required for about the same amount of fun time :)

    --
    .sig: Sique *sigh*
  30. You're the moron, moron. by Anonymous Coward · · Score: 0

    Do you know the difference between "insure" and "ensure"? FFS watch your spelling when you're wailing on someone! It jars even more badly than the usual rediculous[sic] mistakes when you're setting yourself up as some kinda authority...

  31. Even more off topic... by IIH · · Score: 1

    Because when AC and DC is involved, people get confused: It all goes wrong when a radio contestent is asked to spell AC/DC and can't.

    --
    Exigo spamos et dona ferentes
  32. Thay have duel opteron boards with Full X16 SLI by Joe123456 · · Score: 1, Informative

    some of thoes boards have HyperTransport slots, pci-x, pci-e, scsi and sata. Thay sell them as

    High-End PCI-e Graphics (SLI Supported)
    High Performance Gaming Workstation
    http://www.supermicro.com/Aplus/motherboard/Optero n/nForce/H8DCE.cfm

    1. Dual AMD® Opteron(TM) Support, (Dual Core Ready) 1000 MHz HyperTransport Link

    2. nVidia® nForce Pro 2200 (CK804) / nVidia® nForce Pro 2050 (CKIO4) Chipset

    3. Up to 16GB DDR400 SDRAM (or) Up to 16GB DDR333 SDRAM (or) Up to 32GB DDR266 SDRAM

    4. Dual-port Gigabit LAN / Ethernet Controller

    5. 8 SATA ports

    6. 2 (x16) PCI-Express, 2 (x4 using x8 slot) PCI-Express, 3 32-bit 33MHz PCI

    7. AC97 6 channel Audio

    8. 8 Fan support with Speed Control

    H8DCE-HTe is the same with
            1 HyperTransport slot, 2(x16) PCI-Express, 1 (x4 using x8 slot) PCI-Express, 3 32-bit 33MHz PCI

    http://www.supermicro.com/Aplus/motherboard/Optero n/nForce/H8DCi.cfm

    1. Dual AMD® Opteron(TM) Support, (Dual Core Ready) 1000 MHz HyperTransport Link

    2. nVidia® nForce Pro 2200 (CK804) / nVidia® nForce Pro 2050 (CKIO4) / AMD8132 Chipset

    3. Up to 16GB DDR400 SDRAM (or) Up to 32GB DDR333 SDRAM (or) Up to 32GB DDR266 SDRAM

    4. 2 Single-port Gigabit (CK804/IO4) LAN / Ethernet Controller

    5. 4 SATA ports

    6. 2 PCI-Express x16, 1 PCI-Express x4, 2 PCI-X 133/100MHz, 1 PCI-X 100MHz, 1 32-bit 33MHz PCI

    7. AC97 6 channel Audio

    8. 8 Fan support with Speed Control

    http://www.supermicro.com/Aplus/motherboard/Optero n/nForce/H8DC8.cfm

    1. Dual AMD® Opteron(TM) Support, (Dual Core Ready) 1000 MHz HyperTransport Link

    2. nVidia® nForce Pro 2200 (CK804) / nVidia® nForce Pro 2050 (CKIO4) / AMD8132 Chipset

    3. Up to 16GB DDR400 SDRAM (or) Up to 32GB DDR333 SDRAM (or) Up to 32GB DDR266 SDRAM

    4. 2 Single-port Gigabit (CK804/IO4) LAN / Ethernet Controller

    5. 4 SATA ports

    6. 2 PCI-Express x16, 1 PCI-Express x4, 2 PCI-X 133/100MHz, 1 PCI-X 100MHz, 1 32-bit 33MHz PCI

    7. AC97 6 channel Audio

    8. 8 Fan support with Speed Control

    SCSI * Dual Ultra320 SCSI drives with Host RAID * Adaptec AIC-7902W Dual-Channel Controller

    ZCR * Supports Supermicro All-In-One Zero Channel RAID card AOC-LPZCR1 (or) * Adaptec 2010S or Adaptec 2020S

  33. Re:1000 Watts of power!??!?! Not really! by pogson · · Score: 1

    They spec a 1000W supply. The board does not use all that, but they need so many amps at 12V and so on so it takes 1000W supply to handle the load reliably. A typical PC has a 350W supply but only uses 200W fully loaded.
    4 drives X 20W =80W
    4 CPUs X 100W= 400W
    16!!! sticks of RAM X 10W =160W
    mobo itself X 20W = 20W
    Comes to 660W but the power supply and fans use some so I would say 800W would be enough. The extra is just to be safe and to meet peak current loads on every line.

    Reminds me of a hot air soldering gun we used in the old days, like a hair dryer on steroids. The layout is interesting. I wonder if heat pipes or water cooling would be a more sane approach. This would make a beautiful X terminal server. Probably it could handle 300 clients easily. Fully tricked out the cost might be:
    4 dual core CPUs X $1000=$4000
    mobo=$1000
    RAM=$3000
    4 250 gB drives $500
    power supply $500
    Total $9000, about $30/client! A bargain!

    --
    A problem is an opportunity http://mrpogson.com
  34. 1000 Watt power supply, NOT 1000 Watts of power by speedplane · · Score: 3, Informative


    "It also consumes 1000 watts of power."
    WRONG!
    The board requires a 1000 Watt power supply, not neccesarily 1000 watts of power. The power supply is the upper limit of how much the board can consume. Most computers come with a 300 watt power supply even though they normally use only about 100 Watts.
    That being said, this board probably consumes quite a bit of power (but much less than 1000 watts) if it needs such a heavy duty power supply.

    --
    Fast Federal Court and I.T.C. updates
  35. Think about it... by Pranadevil2k · · Score: 1

    Finally.. Nethack in hi-def...

  36. Secret message by this+great+guy · · Score: 1
    <<
    Supermicro has offered AMD solutions for a quite while now - just not under their "main" brand name. If you don't know that their Aplus products exist, you won't find them.
    >>

    That's funny because A plus, in French, means "See you soon". This is probably a disguised message Supermicro is sending to Intel :-)

  37. Optomon by kennygraham · · Score: 1

    And here I thought there were only three of these guys. Ooooh wrong thing.

  38. Dim by Anonymous Coward · · Score: 0

    And all the LED lights in the data centers around the world dim in anticipation.

  39. Re:1000 Watts of power!??!?! Not really! by QuaZar666 · · Score: 1

    while you could you water cooling, it would mean you would have to probally need 1U for the radiator for it. What would probally work better would be fans for the processors, and the controller and then an enclosed Rack with cold air blowing in to the entire Rack.

  40. Re:Imagine what a... by Anonymous Coward · · Score: 1, Funny

    ...beowulf cluster of these would do.

    Run Vista?

  41. Re:Geek logic alert by Anonymous Coward · · Score: 0

    probably more sexually satisfying then your left hand.

  42. Re:Opteron 1xx by BlueBiker · · Score: 1

    All of the original Opteron 1xx and some of the current ones are Socket 940. ISTR reading something about AMD phasing out or increasing prices on 939 Opterons to discourage overclockers.

    Excellent info about the PCI Express slots and lanes. It can be confusing that certain motherboard slots will physically accommodate lane configurations that they don't actually support.

  43. I do this on a smaller scale. by linuxpyro · · Score: 2, Interesting

    I have a VIA mini-itx board that I run as a Web, mail, and MySQL server in my basement. I got the board used, with the 12 volt dc power supply. I use one of these with a small emergency lighting battery for a simple, efficient UPS. The board has a 1 GHZ processor, and draws on average about 3 amps at 12 volts. It should run for about an hour on backup; luckilly I haven't had to test that lately :).

    --
    Saying "I'll probably get modded down for this" in a post is the best way to get it modded up.
  44. Nvidia rocks... by jnelson4765 · · Score: 1

    Especially with driver support. I've seen a lot of machines that are completely unstable with ATI cards that behave properly with Nvidia cards, under Windows. The dual-head support on Nvidia is much more reliable in my experience, as well.

    I would use Nforce 4 mobos for any new AMD system running Windows - and it would be my second choice (after Intel) for the Intel processor machines as well.

    Driver stability is rather hard to acheive (as any Windows admin can tell you), so any company that has a good track record of stable drivers is on my short list of favored suppliers.

    </fanboy>

    --
    Why can't I mod "-1 Idiot"?
  45. How the hell do you cool it? by EmagGeek · · Score: 1

    If you have 8 cores in there along with a full load of RAM and a couple of 10krpm hard disks, how THE HELL do you get all of the heat out?

  46. See the whole server by ZESTA · · Score: 1

    You can see the server platform (Chassis + motherboard) in the following brochure:

    http://www.supermicro.com/manuals/brochure/A+Broch ure.pdf

    -Randy

  47. Supermicro at it again... by Nazo-San · · Score: 1

    Seems this company has been at it a while. My "HTPC" system is an ancient Supermicro board providing a dual Pentium 3 setup backwards compatible (via a jumper to switch the FSB) with Pentium 2 processors. This is back in a time where dual processor was almost server only, yet, this solution was affordable enough that we could get several of these for various people, one including myself, and useful enough that I haven't thrown away the thing yet today (I still use it as a server as well.) I'm not saying they are pioneers or anything, but, they know what they are doing and have practice at it.

  48. electricity cost may approach $75/month by MMHere · · Score: 1

    Let's say your electricity costs you $0.10 per kilowatt-hour. This machine will cost nearly $75/month to operate:

    1000watts / 1000 = 1kW-hour to operate for one hour

    ( 1 kw-hour * 24hours per day * 365 days / 12 months ) * $0.10 == $73.00 / month

    1. Re:electricity cost may approach $75/month by Anonymous Coward · · Score: 0

      Compared to how much to operate two SMP or four UP machines?

    2. Re:electricity cost may approach $75/month by Nazo-San · · Score: 1

      Well, I just wanted to clarify a bit on my earlier comment since I posted it as I was stepping out the door. Firstly, I'd like to point out that while manufacturers will say that so-and-so is required, this is not always so. I have heard of people running setups that claim to require 350W minimum on 300W and I think one rare case even 250W and somehow it was stable. Mind you, they had onboard video, the cpu, memory, and FSB were stock clocks, and so on (one thing to bear in mind is that Athlon 64/Opteron is built on mobile technology, so is quite power efficient.) As for how you get 1kW, you might be able to use two or three normal PSUs instead of one uber PSU. I have heard of setups doing this, albiet usually for a serious harddrive array (one example of which had so many HDs they had to create a seperate tower just for them.)

      BTW, one thing you'll want to bear in mind is that a x wattage power supply does not use x watts. Firstly, none are 100% efficiency (which is why some use a PR rating saying the number of watts they REALLY do instead of lying and saying how many they can do in a vaccuum with carefully stored battery generated power instead of real life power sockets.) That's the first half of my point. The second half is that they don't use that much more than what they need at a given time. If the system is drawing 200 watts of power, that's what is being consumed. It's not consuming 1000 watts of power at all times irregardless of system usage. In other words, every single time those processors hit an idle cycle and decrease power usage, every time the memory enters a wait state, every time the harddrives stop running non-stop, power draw is going to go down. Supermicro covered their bases in a worst case scenario where you're running four copies of Prime95 each set to write their log files as quickly as they can to a raid array because you have to assume the worst scenario for a server. In real life, like someone said earlier, you're probably using 660W of power or so. Let's say 700. At 0.10 per kWh, that's ((0.7*24*365)*0.10) / 12 = 51.1.

      Actually, one advantage here is that you can focus your cooling a bit more. A very well aired out case with four processors will actually run cooler than two cases with two processors each or especially four cases with one each. When combining a large numbers of these systems in a single room it may not run any cooler by itself mind you, but, it shouldn't run hotter either by itself. However, if you have only a few of them, it becomes possible to focus your AC or even buy refridgerated cases (yes, there is such a thing, or at least was, I remember seeing one primarily intended for servers a few years ago. The price was insane, but, then it would be more reliable than even watercooling for a server situation.) Anyway, run your components cooler, and they actually consume a little less power (which adds up in a hurry in a system that's on 24/7.) Of course, a poorly designed case would run even hotter, so it does rely on the use of a little sense.

  49. drive spinup by hawk · · Score: 1

    The original MacPortables (including backlit) could not power up without their battery for this very reason. The 1.5A power supply wasn't enough to initially spin the hard disk. If you used a later 2.0A or 2.5A powerbooko supply, they could start without the battery.

    hawk, who still has his