AMD Packs Six-Core Opteron Inside 40 Watts 181
adeelarshad82 writes "Advanced Micro Devices has launched a low-power version of its six-core Opteron processor in time for VMworld, a key virtualization show that opens on Monday. The six-core AMD Opteron EE consumes 40 watts, and is designed for 2P servers, among the most popular in the virtualized server space."
Hardware (Score:5, Funny)
The six-core AMD Opteron EE...is designed for 2P servers...
All I really want to know is: can you install it in a toaster?
Re: (Score:2)
Re:Hardware (Score:5, Informative)
What reputation? Since the days of the original Thunderbird core (which still ran cooler than comparable P4s, though admittedly didn't have meltdown prevention circuitry), AMD has consistently given Intel a run for their money in that regard.
Now, the Atom has finally brought Intel back to the realm of "reasonable", but it doesn't seriously compete with AMD, it competes with VIA (and poorly at that - The Nano blows the Atom away, clock for clock and Watt for Watt).
Don't get me wrong, Intel has certainly regained my respect when it comes to performance, but to call AMD the toaster requires ignoring the past 10 years.
Re: (Score:3, Interesting)
Re: (Score:3, Interesting)
Re: (Score:3, Interesting)
Re: (Score:2)
Re: (Score:2)
The K6 series were very hot compared to Intel equivalents.
Irrelevent to his argument. See below:
Don't get me wrong, Intel has certainly regained my respect when it comes to performance, but to call AMD the toaster requires ignoring the past 10 years.
Why yes, time flew by that quickly.
Re: (Score:3, Interesting)
Indeed. It's been a very long time since AMD has been the hotter running CPU. It was Intel that introduced us to heatsinks that could hurt you if you dropped them on your toes.
It's hard to believe that at one time CPUs didn't have heat sinks at all.
Re: (Score:2)
It's hard to believe that at one time CPUs didn't have heat sinks at all.
Hehe. Yeah, and I remember getting bit by the change and feeling rather foolish. I'd built a computer for my step-sister, and, well, I really never even thought about adding a heat sink. About a year and a half, maybe two years later, she was complaining that it kept crashing so she was just going to the computer lab (defeating the point of her having her own), so I took it and checked it out. Well by this point my own system very
Re: (Score:2)
I was amazed the first time I saw a CPU with a loose heatsink where the label on the bottom actually charred. Alas, it became a common diagnosis for AMD. Intel overheated as well, but I tended to diagnose that by the machine suddenly slowing down to a crawl.
I am glad AMD added internal thermal monitoring. Before that they weren't the hottest running, but did suffer more death by heat.
Re: (Score:2)
I am glad AMD added internal thermal monitoring. Before that they weren't the hottest running, but did suffer more death by heat.
No kidding! It was way too long to get that going. If they'd had it from the beginning, I probably would have only been mildly embarrassed when my processor didn't boot and I figured out I hadn't taken the sticker off the bottom of the heatsink, rather than extremely humiliated by my own noobness and a $200 processor-turned-shitty-paperweight. :)
Re: (Score:3, Interesting)
AMD was extremely sloppy on power management before the K8/Opteron days.
See my old /. Journal on the subject: http://slashdot.org/~evilviper/journal/70512 [slashdot.org]
In short, while the maximum power of AMD CPUs was about the same as their P-III equivalents, AMD chips (Thunderbird
Re: (Score:2)
With AMD's reputation for producing hot-running processors
What reputation?
You're posting on a site that still mods BSOD jokes as funny.
Re: (Score:3, Insightful)
>>>With AMD's reputation for producing hot-running processors ...
Don't you mean Intel? After all their early 90s Pentiums were the first CPUs to spark the "you could fry an egg" jokes. And the Pentium 4 sitting in my computer is a major power hog (~90 watts), and it's just a single core.
Anyway 40 watts for six processors isn't really that bad. About 7 watts each.
Re: (Score:3, Funny)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
can you install it in a toaster? Yes, but at only 40 watts, it will take forever to brown your bread!
He only needs to replace NetBSD in his toaster by Vista and to overclock it slightly.
Re: (Score:2)
Not a good idea... (Score:5, Interesting)
If you were willing to deal with the size and weight of those high-end gamer laptops, the ones with quad core i7s and SLI, you could probably build a 17-inch dual socket system....
Wouldn't be a laptop I'd want to use (Score:4, Informative)
Now if only someone would wise up and build a 15" laptop with an Atom chip, and LED display and a 9-cell battery... mmm, 8+ hours of battery life.
Re: (Score:2)
Now if only someone would wise up and build a 15" laptop with an Atom chip, and LED display and a 9-cell battery... mmm, 8+ hours of battery life.
They can't. Intel doesn't allow the Atom in full-size laptops at all, or something stupid along those lines.
Re: (Score:2)
Re: (Score:3, Interesting)
Dual UltraSPARCs, up to 16 gigs of RAM, full sized 64 bit PCI slot, 3GbE ports. Of course, it's 22 pounds, and they don't even say what it costs.
Re: (Score:2)
Re: (Score:3, Insightful)
What do you need that extra speed for?
less time wasted staring at an hourglass cursor, of course.. atom is dog-slow and it might be fine for these netbook toy-laptops but I would never buy something resembling a full-sized laptop with a dinky atom processor in it. it's barely tolerable with firefox and xp.
if you aren't seriously bothered by the speed of atom processors then i will assume you run linux/fluxbox (or the equivalent) and a very lightweight web browser. and if that's the case (you're already running an OS that is not restricted
Re: (Score:3, Informative)
TFS is a bit light on details (Score:5, Informative)
Here are a few quick bits from the article:
Re:TFS is a bit light on details (Score:4, Interesting)
Compare to the 2377 EE, 40-watt quad-core @ 2.3 GHz: approximately 1/3 more performance from the new six-core chip.
Depends on what kind of server. If you're talking about a Web server, IIS 5.1 and later or Apache 2.x and better with multithreading on, yes. If you're talking about Apache 1.x or 2.x without multithreading, or some older versions of IIS, no.
Re: (Score:2)
Interesting; straight math didn't seem to support the article's claim of a 1/3 performance gain, but I assumed the increased parallel capability must be responsible for the extra performance. I'm glad someone with more knowledge was able to clear that up.
Re:TFS is a bit light on details (Score:5, Informative)
Re: (Score:3, Interesting)
Except that with multiprocess concurrency (i.e. non-multithreaded Apache on Unix), you actually gain in a NUMA setup like the Opterons have been from day 1. See, in the optimum case in a NUMA environment, the server process that handles a request gets an entire memory bus for itself. That's far more scalability than with multithreading in the absence of memory duplication, which AFAIR Linux doesn't implement on a per-thread basis in the same address space.
This is why Opterons practically own the 4-socket x8
Re: (Score:2, Interesting)
So your comment ignores the fact that this CPU will probably be running 6 (or more) VMs, which could just as well run single-threaded code....
Clearly that's the market that this chip is targetting. I'm simply pointing out that 1) even if you're not in that space, this chip compares favorably to a 2.3 Ghz Quad Core or even a 3.2 Ghz Dual Core, so long as you're running multithreaded apps on any OS that uses a sane threading model. Just understand that single-threaded apps won't compare unless you're running them virtualized.
OTOH, separate blades are going to give you to better performance, no matter how you look at it. On the gripping hand, if
Re: (Score:2)
Re:TFS is a bit light on details (Score:5, Interesting)
The important information FTFA is here:
"AMD also estimated that the power consumption for a fully populated 42U rack would be 9.2 KW using the six-core Opteron 2425 HE, a 55-W part. Replacing those chips with the 2419 EE would require 7.5 KW, about an 18 percent power savings."
That's just in the rack consumption. I would imagine these probably run cooler, too, which will help with HVAC costs.
AMD seems to be doing a better job shrinking down dated designs at this point. While Intel is selling the Atom, which is undoubtedly cooler and less power-hungry, it's still based on a very old CPU design, which isn't up to heavy computing tasks. AMD, OTOH, has now established a pretty good record of taking mainline processors, and developing lower-power versions. They scaled down what used to be a pretty hot Athlon core (Thunderbird) to the Geode (as used in the OLPC). They followed that with a 45W Athlon 64 X2. Now the Opteron. Intel does have a 35W Conroe, but it's in Celeron cripple-mode badging, a shadow performance-wise, of the original C2Ds that initially came out on that core.
I hope that AMD does release a desktop version of this, but I don't know if they could keep it profitable ($900+ eek.)
Re: (Score:3, Informative)
Actually the geode used in olpc is not a scaled down thunderbird. The Geode LX series is based on old cyrix chips (although the Geode NX is a scaled down athlon).
Cyrix (Score:2)
I running a funky AMD four core with a TLB bug. Works fine on Windows XP. I own Intel stock but use AMD chips. I'm looking forward to using one of these low power chips on a HTPC.
Re: (Score:3, Interesting)
> That's just in the rack consumption. I would imagine these probably run cooler, too, which will help with HVAC costs.
I understand that for every "power watt," it takes 1-2 additional "cooling watts" additional power, in a server room.
So, if a rack takes 10KW, expect an additional 10-20KW of electricity to cool the server room.
I'd, then, estimate 30KW total for a 10KW rack, just to be safe.
So, an 18% savings on 10KW (1.8KW saved), is really saving you on the order of 3.6KW to 5.4KW, when you include co
Re: (Score:2, Insightful)
I love how AMD is touting the lack of DDR3 support on a new chip as a "feature".
Re: (Score:3, Insightful)
I love how AMD is touting the lack of DDR3 support on a new chip as a "feature".
That was my initial reaction too. However, remember that most (not all, obviously) typical server tasks aren't particularly memory bandwidth-hungry. Email, web serving... even databases aren't usually coming anywhere close to saturating the bandwidth DDR2 can provide, even with several virtualized OSes sharing that bandwidth.
Re: (Score:2)
Re: (Score:2)
2P (Score:5, Funny)
Do they mean Dual Processor? I've never heard the term 2P server before.
Re: (Score:2, Insightful)
Re: (Score:2)
Re:2P ... (Score:5, Funny)
or not 2P, that is the question.
It might get unpleasant if you hold it in too long.
Re: (Score:2)
Schrodinger's [angryflower.com] beehive -- two bees, or not two bees? THAT is the question!
2P or not 2P is a bad question. As General Patton said (at least in the movie portrayal of him), "never turn down a chance to piss!"
Re: (Score:2, Insightful)
Do they mean Dual Processor? I've never heard the term 2P server before.
My guess would be 2-partition (as in, two virtual partitions on a single physical server).
Re: (Score:2)
I've never heard it before either. Seems like a SEO-only term. You see it in the head tag, but the marketing dribble doesn't mention it. Server specs for anything coming up "2p" includes 2 processors, but correlation != causation, especially with marketing folk.
2P = Dual Socket (Score:2, Informative)
Single socket (1P), Dual socket (2P).
Re: (Score:2, Informative)
Apparently, "2P" does indeed mean dual processor in adspeak.
Citation: http://searchoracle.techtarget.com/generic/0,295582,sid41_gci1362417,00.html [techtarget.com]
Re:2P (Score:5, Informative)
I've seen it before, usually used in a context where you have 2P, 4P, 8P = dual-processor, quad-processor, octo-processor machines because noone wants to go around remembering what that should be abbreviated like. Of course, with cores per chip varying widely just saying you have a DP/2P machine says little these days.
Re: (Score:3, Insightful)
Of course, with cores per chip varying widely just saying you have a DP/2P machine says little these days.
Naw, it's just been transmuted to mean processor sockets. Which from a system architecture standpoint is the more meaningful way to do it. You can put anything from a single-core to a six-core processor in a given socket (assuming they all exist in the necessary package), but you can't change the number of sockets in your motherboard. So "2P server" tells you that there are two sockets which you coul
Re: (Score:2)
Works fine until we start getting specialised cores with specialised instructions. I've adopted sockets and processors. A pizza box can 8 processors (two sockets). A single processor, in this case, is still a generalised processing element.
Re: (Score:2)
Works fine until we start getting specialised cores with specialised instructions.
I don't see why heterogeneity should change anything.
The processor on my desk contains two x86 cores, and a 'north bridge' memory controller with DMA engine and APIC. Soon there will be ones that also have a graphics core on them as well. I'd still call that a processor.
The Cell Processor contains a single general-purpose Power core, 8 relatively specialized cores, and a memory controller, but I feel comfortable calling the
Re: (Score:2)
I see the point you're making, but what happens when I write a parallel program? I make a request to the array services daemon that I want 4 processors and that gives me a single socket (in my current, massively outdated system :) ), or in your nomenclature a single processor. I see it as having 4 processors (or cores if you will) and hence the need to get a naming scheme that is consistant across hardware and software.
And please don't say threads :s
Re: (Score:2)
I see the point you're making, but what happens when I write a parallel program? [snip] I see it as having 4 processors (or cores if you will) and hence the need to get a naming scheme that is consistant across hardware and software.
And please don't say threads :s
Well sorry, I hate to break it to you, but regardless of whether you use my nomenclature or yours, that's the only term that is going to be consistent across hardware and software.
Take for example a case I already mentioned: The Nehalem (Core i7)
Re: (Score:2)
Re: (Score:2)
I've seen it many times in UK confectionery stores.
Gaming/compiler performance? (Score:2)
Re:Gaming/compiler performance? (Score:5, Informative)
6 x 1.8 = 10.8
2 x 3.2 = 6.4
If you can take full advantage of the six cores, there's a lot more computational power despite the slower clock speed.
Re: (Score:3)
Analogy fail. It's more like saying 6 assembly lines producing 1.8 cars per hour can make more cars per hour than 2 assembly lines producing 3.2 cars per hour.
Assuming you have stuff pipelined correctly so as to take full advantage of all six cores, of course... which I explicitly required in my previous post. If you're only able to keep 2 cores reasonably busy, the 6-core processor will be slower because 4 of its cores are mostly idle.
Re: (Score:2)
While I don't disagree, I'd like to point out that you described situations where the 6-core processor can't be taxed to its full capability.
In other words, your main point was that the only situation where it can be fully loaded is "in the server market [and] aimed entirely for vm where the multiple cores will help the most". That's a valid point, but it's not presented foremost in your post. It's sort of hidden down at the bottom.
Re: (Score:3, Informative)
Re:Gaming/compiler performance? (Score:5, Informative)
Also, by shoehorning this into a 40w envelope, they're obviously going for power efficiency over horsepower. Interesting fact: power usage is one of the largest costs of a data center, and it's growing.
Re: (Score:2, Funny)
Now I finally understand why MUD was banned on our group server.
Re: (Score:3, Informative)
All about threads (Score:2)
This is for situations where you need lots of processes running but that those processes are either easily completed, are low-impact, or limited by bandwidth or the user. Web servers love lots of cores
On gaming you could separate the game into a user environment thread, a physics thread, an object management thread, a pair of AI threads, and still have a core left over for general OS activity.
I know that in theory compilers could also pull loops and modules out to separate threads but I haven't the foggiest
Re:Gaming/compiler performance? (Score:4, Interesting)
I had this argument with someone once. They didn't quite get it. The machine they were using was a 4 CPU 700Mhz server. In their logic, 700Mhz * 4 = 2.8Ghz. I wanted to move them to a 2 CPU 1.4Ghz machine, which I promised would be blazing fast. In their mind 1.4Ghz * 2 = 2.8Ghz, so there was no difference.
There were a bunch of reasons for the move. The hardware was old. The form was huge (like 5u tall) and power hungry. The OS needed to be updated badly, and we couldn't take it offline for a day to do that. One day there was a fault of some kind (it's been a while, I don't remember specifically), so we moved it over to the new machine that I had wanted to move them to. They were amazed. Their $40,000 server had been replaced by a $2,000 server (original costs for both), and it was running faster and better than before. After the move, I repaired their old server, upgraded the OS, and made it ready. I offered to move them back, and they refused. :)
About a year later, we had a 2CPU 2.4Ghz machine ready for them, and I offered again, "May I move you?" This time there wasn't a complaint. We just scheduled a window and did it. I set a 3 hour window, and we had it completed in about 15 minutes.
I agree, I'd rather have CPU speed AND cores. I'd sacrifice extra cores for more speed. CPU speed has stagnated, while they're growing cores. I remember this happening in the past too, around the time CPU's were 200Mhz. You could get motherboards that supported one CPU, then 2 CPU, then 4 CPU, but the speeds weren't going up. You could give me 100 CPU's at 200Mhz, but I'd rather have one at 10Ghz.
I'm sure people will throw a bunch of excuses of why. I remember back when the 50Mhz CPU was the fastest available, there were all kinds of reasons thrown around of why CPU's would "never be faster". People were very insistent that they were right. There were RF interference issues. If CPU's got to RF speeds, radio and TV would cease to work. If we got up near 2.4Ghz, people would be cooked because it's the same frequency as microwave ovens. There was no way to deal with the thermal issues, and computers would be ovens requiring liquid cooling (like liquid nitrogen or helium, not water cooling). Blah, blah, blah, blah. As we've seen, we did get well beyond 50Mhz. It's just a matter of time. I'm just disappointed that we end up stagnating. It's probably financial issues. The market will support a slower multicore CPU, but people won't spend the money on faster CPU's right now.
I always love the "latest greatest" craze. It's entertaining. People will spend mad money on latest greatest, and I'll wait 6 months or a year to buy the same thing at a fraction of the cost. Maybe I'm part of the problem there. I won't drop $500 on a CPU, but I'll drop $100 on last years model that's only slightly slower.
At least right now it's nice, since I can buy older and older hardware, and really not be far behind the curve. :)
Re: (Score:3, Insightful)
> I always love the "latest greatest" craze. It's entertaining. People will spend mad money on latest greatest, and I'll wait 6 months or a year to buy the same thing at a fraction of the cost. Maybe I'm part of the problem there. I won't drop $500 on a CPU, but I'll drop $100 on last years model that's only slightly slower.
Say you wait 6 months before upgrading. $400 divided by 6 months = $2 per day.
Assuming you earn $20 an hour, if the new chip saved you just 6 minutes per day, then it's a worth while
Re:Gaming/compiler performance? (Score:4, Insightful)
For the one paying the wages, there certainly can be a savings. So, for a company that is paying an employee, your math can be correct in some instances.
That all being said, from a non-economic standing, it may still make sense to upgrade. I know, I would rather have the extra 6 minutes of time, even if it is just spent getting a cup of coffee, or just being productive on something else. Ok, Ok, even if it is spent posting on Slashdot about how I would rather have the extra 6 minutes.
Re: (Score:2)
I wait until the latest and greatest has gained the larger part of retail shelf space then buy the most established and robustly tested stuff at the price nearest it's exit price just before retailers stop carrying it. The problem with the latest and greatest stuff is that often there's bells and whistles that won't make it because the market goes in another direction.
I just built an intel Quad core for under $400, but used some components from cannibalized systems. Right now USB, SATA, and the PCI bus all
Re: (Score:2)
While I do agree, I've been considering replacing my desktop (and adding a second one), when money allows. CompUSA is still alive in my city, so I wander the isles when I'm there. I can get a $300 machine that's faster than any of my old desktops. Ok, so it's an eMachines, big deal. I bought one about 2 years ago to be a mail filter server. It was an AMD64 with 1Gb RAM. I upgraded the RAM immediately (we needed more for the mail load). The motherboard died, so I picked up an Asus board
Re: (Score:2)
Re:Gaming/compiler performance? (Score:5, Informative)
CPU speed has stagnated
It hasn't stagnated at all. You're equating cycle rate with performance, that's incorrect.
Each processor architecture does a different amount of work each cycle. Counting only the number of cycles is like comparing the running speed of two men by the number of steps they take each minute - but one guy may be a midget and the other eight feet tall. Clock speeds remain similar but performance doesn't correlate.
For example, a 3Ghz P4 isn't even half as fast as one core from a 3Ghz Core i7. The number of instructions per clock have been continuously improving with each new architecture.
Phenom is faster than Athlon X2. Phenom II is faster than Phenom.
Core 2 is faster than Pentium 4. Core i7 is faster than Core 2.
So you can have what you want - improvement continues in both per-core performance and the number of cores.
Was it the processors or the memory? (Score:2)
Re: (Score:2)
I won't argue the memory probably helped with the overall speed, just as better drives did, but all in all I'd say it was the CPU speed. The faster memory and better drives are an added bonus. :)
It was really hard to get him to step away from that old server, since there was such an investment made in it. In the end, after we decided not to put that server back into production for that purpose, I couldn't convince anyone to use it. I don't know it's final fate, but I did sug
Re: (Score:2)
One possible way to solve this is to figure out how to scale single processes over multiple cores. As far as I know, that hasn't been done yet, and I don't even know if it's possible. But if it is possible, that adaptability seems like it'd kick the whole "programming for multiple cores" thing in the ass pretty readily.
Multithreading? Check, we've got that. Multiprocess? Sure, we've got that too, though there are still a lot of applications which don't do it, or do it wrong. But if it could be done transpar
Re: (Score:2)
CPU speed has stagnated,
Not even close. Floating point performance on a single core has still been increasing according to Moores law. 256 bit wide registers, vectorised/packed instructions, xsse4.1, etc
Re: (Score:2)
Re: (Score:2)
WTF is 2P? (Score:2)
How many U to a P?
Or is that supposed to be dual CPU/dual socket?
Re: (Score:2)
Re: (Score:2)
Shouldn't T to A be 2:1?
Good for sore muscles (Score:2)
This makes me feel all warm and toasty inside.
My electric heating pad, which helps me with little muscular issues, is 50 watts, but that dissipation is spread out over a 30 cm x 65 cm surface.
What Does Intel Have... (Score:2)
We're getting robbed of latin related fun (Score:4, Funny)
The previous generations of multi-core CPUs weren't 2-core and 4-core, they were dual and quad-core. These new chips should pretty obviously be called sex-cores. Not since the 667MHZ PII have I been so disappointed.
Re: (Score:2)
5th core
Re: (Score:2)
6th Core, but I'm the bad one so the 7th core will have to be the 6th one.
Re: (Score:2)
Re: (Score:2)
2 Processor
Re: (Score:2)
Re: (Score:2)
Re:Gimme MHz (Score:5, Informative)
That worked great for the Pentium 4, didn't it? Faster clock != more instructions per second. The only way to get close to 4GHz on the Pentium was with a 31-stage pipeline. http://en.wikipedia.org/wiki/Instruction_pipeline [wikipedia.org]
This means, on an instruction like if(a+b>c){}, the actual branch gets delayed by about 20 cycles if the processor guesses incorrectly whether the if statement should execute or not. Add the overhead due to such a fast clock (the P4 could only have 4 logic gates per pipeline stage due to the speed).
I'll keep my more efficient, better laid out processors over raw GHz, thank you very much.
Re: (Score:2)
Re: (Score:3, Informative)
The P4 only managed 4 gates at 4GHz because it was 90nm manufacturing. Phenom II and the i7 are 45nm, and the faster gates enable the higher clock speeds naturally and without huge tradeoffs, unlike the P4 where GHz drove development instead of performance.
Re: (Score:2)
My Phenom II 955 will even do 3.6GHz on the stock cooler now, probably even a lower thermal envelope than the P4's of the time. And the Core i7's (I'm too cheap for a full system replacement) are faster and cooler still.
To add to my sibling post, the P4 was supposed to be at ~10GHz by now. Intel's projections based on further technology shrinks, and further extended pipelines, was pretty reasonable from a transistor speed and clock speed/IPC tradeoff perspective. They failed to account for an explosion in
Re: (Score:2)
Except for the "wah!" there's truth to what you say. Well, excluding the "give me death" part as well. In fact, let's just concentrate on the middle bit.
It would be nice to see some greater increase in speed. There are reasons to have that. But in the meantime, I will accept code properly written to utilise multiple cores as a substitute.
Re: (Score:3, Informative)
$1000 for the processor is peanuts for certain applications. Generally for something like a database server most of the cost goes into software anyways (assuming you're not using an OSS database - if you are that's fine and some of this doesn't apply, but my employer simply doesn't allow it). With MS SQL Server for example, if you're not licensing by CAL's (which is a budgeting headache) is licensed per processor - at about $5000 or so per processor. Neat thing though is multiple cores don't count - only