Slashdot Mirror
AMD Packs Six-Core Opteron Inside 40 Watts
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."
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?
But with a 40 watt chip you could get that into a laptop, if you felt like it. Not the thinnest, lightest, or quietest laptop around; but plenty of 14-15 inch units under two inches thick(though often not far under) were running P4s at least that power hungry back before P-Ms became cheap enough for common use.
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....
Here are a few quick bits from the article:
Alexander Peter Kristopeit bought his basement from his mommy for one dollar.
Do they mean Dual Processor? I've never heard the term 2P server before.
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.
Alexander Peter Kristopeit bought his basement from his mommy for one dollar.
From TFA: "According to IDC data quoted by Brent Kerby, a product manager for the chip, about 82 percent of cloud and Web servers only use about half of their available processor power at any given time." Not intended for gaming or compiling. Low power, multiple cores, it's a server chip.
or not 2P, that is the question.
It might get unpleasant if you hold it in too long.
Most laptops today have much more power efficient chips (AMD's line tops out at 35W, Intel's 25W, most do quite a bit less, especially with all of the fancy power-saving junk thrown in like QuickStart and SpeedStep w/ deeper-sleep DC4). And both of those numbers are just embarrassing with chips like the newer dual-core Atom chips which run at 4W or less at full-tilt and do most everything anyone demands of a laptop anyways.
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.
"Victory means exit strategy, and it's important for the President to explain to us what the exit strategy is." G.W.Bush
This is a server processor. If you are either gaming or compiling on your server, you are doing something wrong. My servers here at work tend to do a high volume of low processor intensity transactions... therefore, more cores (and more simultanious transactions) is far more important than high speed.
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.
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. :)
Serious? Seriousness is well above my pay grade.
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 you have 10 blades, each running this 6 core CPU (conceivable in 40 watts!) and a VM hypervisor... sounds delicious.
My blog
Single socket (1P), Dual socket (2P).
> 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 investment to upgrade sooner rather than later.
Now I finally understand why MUD was banned on our group server.
I like your math, but you do have to change it just a tad. The $2 savings is only if he is paying for the work. If he is getting paid, then there is no savings in getting the job done 6 minutes sooner. If he is hourly, he will get paid $2 less each day, while spending an extra $2. This means a $4 a day loss. If he is salary, then there is no change in his income, but he still pays out the $2 a day in equipment costs, and thus still loses money.
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.
Power usage is actually going down per unit work, by a LOT due to virtualization. Compared to my standard server from just 3 years ago I can do 17:1 virtualization today without any major over-subscription of resources.
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
$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 an actual physical processor counts, so for licensing reasons the more cores they can pack onto a single chip, the cheaper my licensing becomes. I've been setting up dual quad-cores on most systems lately just to keep the official processor count at 2, but six-cores would help even more. Remember too that most decent servers will be running SCSI/SAS drives in a RAID config, and many will also have fairly large tape drives (an LTO3 or 4 will up the cost another few thousand by itself).
With those prices, most of the servers I've setup recently have run $15k-$25k with software. None of them are going to get those $50 Celeron's anyways, so these chips generally won't up the cost that much.
"People who think they know everything are very annoying to those of us who do."-Mark Twain
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.
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
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.
Write your representatives! Repeal the 2nd Law of Thermodynamics!
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.
Fsck the millennium, we want it now.
Millennium Crisis Line: 0890 900 2000 [calls cost 50p/min]
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.
Alexander Peter Kristopeit bought his basement from his mommy for one dollar.
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.
Write your representatives! Repeal the 2nd Law of Thermodynamics!