AMD Launches Piledriver-based Opteron 6300 Server Chips

Subject: Processors | November 6, 2012 - 01:15 PM |
Tagged: server, piledriver, opteron, datacenter, cpu, amd

AMD announced new server processors on Monday based on the same Piledriver architecture used in the Trinity APUs and Vishera desktop CPUs we recently reviewed. With the release of the Opteron 6300 series, AMD is bringing Piledriver to the server room.

The new chips – similar to the desktop counterparts – bring several performance improvements over the previous generation 6200 series Opterons based on the Bulldozer architecture. AMD is positioning the chips as a upgrade path to existing servers and on merits of performance-per-dollar efficiency. As is AMD's fashion, the new chips are competitively priced and "good enough" performance-wise. With 6300, AMD has stated the goal is to reduce the TCO, or Total Cost of Ownership for servers used in data centers, supercomputers, and enterprises by being compatible with existing AMD server platforms with a BIOS upgrade and representing efficiency improvements over previous chips.

Opteron_6300_Series_hand_metallic_background.jpg

The Opteron 6300 series CPUs themselves build upon the Vishera desktop parts by adding more cores and more L3 cache. The server parts will have up to 16 cores clocked at 2.8GHz base and 3.2GHz turbo. They will have TDP ratings between 85W and 140W and will feature prices from $500 to $1,400. On the cache front, the chips have a 16KB L1 data cache per core, 64KB L1 instruction cache per module, 1MB L2 cache per core, and a shared 16MB cache per socket. AMD has included a quad channel memory controller that supports DDR3 up to 1866 MHz and 1.5TB per server in 4P configurations. AMD has rounded out the chips with four x16 HyperTransport 3.0 links rated at 6.4 GT/s per link. Up to 4 processors per server will be supported, which means a maximum of 64 cores.

Opteron_6300_die_shot_16_core.jpg

With Piledriver, AMD added a number of new instructions including FMA3, BMI, and F16c. The company has also implemented server tweaks to the Bulldozer design to improve branch prediction, instructions per clock, scheduling, and reduced the power draw at higher clockspeeds allowing for the chps to clock higher while staying within the same power envelope of the Bulldozer-based Opteron 6200 series.

AMD is using the same socket as the 6200 series processors, and the new chips can be deployed as an upgrade to the old servers without needing a new motherboard.

Screenshot (339).png

When pitting the new Opteron 6380 to the previous-generation 6278, AMD is claiming a number of performance increases, including a 24-percent and 40-percent improvement in SPECjob2005 and SPECpower_ssj2008 respectively.

Further, the company is claiming competitive performance in server workloads with the Intel competition. AMD offers up benchmarks showing the Opteron 6380 and Xeon E5-2690 trading wins, with the AMD part being slower in the STREAM benchmark, but being slightly faster in LAMPS and NAMD. The allure of the Opteron, according to AMD is that the AMD part is almost half the price of the Intel processor, and is hoping the lower priced parts will encourage adoption. AMD argues that the money saved could easily go towards more RAM or more storage (or simply be saved of course).

Screenshot (338).png

The company has announced that its first major design win is Big Red II supercomputer at Indiana University. Built by Cray, the Big Red II will feature 21,000+ Opteron 6300-series CPU cores paired with NVIDIA GPUs. It represents a massive increase in computing power over IU’s previous Big Red supercomputer with 4,100 CPU cores, and will be used for medical, physics, chemistry, and climate research. Beyond that, AMD has stated more that 30 hardware vendors are slated to introduce servers based on the new Piledriver-based Opteron processors including HP, Dell, Cray, SGI, Supermicro, Sugon, and (of course) SeaMicro. On the software side of things, AMD is working with Microsoft, VMware, Xen, Red Hat, and Openstack. The company also stated that it is leaning on the experience and knowledge gained from the HSA Foundation to improve software support and guide the future direction of Opteron development.

Screenshot (336).png

The Opteron 6300 series is an interesting release that brings several improvements to the company’s server chip offerings. At launch, there are 10 processors to choose from, ranging from the quad core 6308 clocked at 3.5GHz for $501 to the top-end 6386 SE with 16 cores (2.8GHz base, 3.5GHz max turbo) and a $1,392 price tag. The 6366HE is an interesting part as well. It is the same price as the 12-core, 115W TDP Opteron 6348, but its has 16 lower-clocked cores and an 85W TDP. With the non-HE edition processors with 16 cores starting at $703, the 6366HE for $575 is a decent deal if you need multi-threading more than a fewer number of higher clocked cores.

Another bit that I found intriguing is that in a few years, AMD will (likely, if all goes according to plan) be offering processors for just about every type of server. They will have low cost, low power ARM Cortex-A57 based chips, Accelerated Processing Units (APUs) well suited to mixed workloads including GPU-accelerated tasks, and CPU-only chips with lots of traditional x86-64 cores. It seems that Intel will continue to hold the high end on pure performance, but AMD and its SeaMicro server division have not given up competing in the server room by a long shot.

[

Further reading:

The Piledrive architecture and Vishera desktop CPU review and The future of AMD: Vishera and Beyond at PC Perspective.

Intel Wants To See 48-Core Processors In Future Smartphones

Subject: Processors | October 31, 2012 - 02:25 PM |
Tagged: smartphone, R&D, prototype, Intel

Intel researchers located at the company’s Barcelona, Spain labs are working on multi-core solutions. Specifically, the researchers are looking to bring massive numbers of processing cores to smartphones and tablets. The x86 chip giant hopes to see as many as 48 core processors powering mobile devices within the next five to ten years.

Currently, quad core System on a Chip (SoC) processors represent the highest number of cores in a mobile device, so a 48 core processor would be an absolutely massive jump. Of course, there are several issues that Intel will need to address in order for such a chip to be feasible. The package size, TDP, and power draw will all need to be drastically reduced in order to fit into the power envelope and form factor of smartphones in particular. The biggest issue standing in the way of such a chip though is software. Massively multi-threaded software is still extremely rare, and on mobile devices is no where close to effectively utilizing the number of cores Intel wants to provide.

Intel Prototype Chip.jpg

Image credit: Computer Wold. A prototyping platform running software to research efficient multi-core processing.

Analyst Patrick Moorhead argues that five to ten years is effectively an eternity in technology-time, and by the time the hardware with massive numbers of cores is feasible, the software will be there. Intel is a bit less optimistic, but hopeful that developers will embrace the idea of multiple low power cores versus a few high clocked, power hungry cores. Such a transition in the software industry would allow smartphones to multitask much more efficiently than they do now, and would no longer have to comprise on the extent of background tasks in order to keep the user interface stable and snappy.

Intel is currently using cloud computers to analyze multi-threading and what tasks can be parallelized. The company envisions being able to encrypt email, listen to media, sync with backup services, and use voice recognition without needing to reach out to Internet-connected servers at the same time. Cores would be able to split the work or dedicate certain numbers of cores to tasks like the UI or video playback. Intel has already demonstrated the ability to turn off unused portions of the CPU to save power as well as its turbo boost modes to increase clockspeeds when there is TDP headroom. It will be interesting to see a 48 core chip, especially if software developers can be coerced into doing tasks with massive numbers of cores in mind.

Interestingly, AMD is going for heterogeneous cores (CPU cores, GPU cores, ARM cores, etc) while Intel is using its process node and chip technology lead to throw large numbers of homogenous cores at the problem of processing. In the end though, it all comes down to battery tech and software enabling these kinds of advancements. I'm hopeful that I'll see these kinds of currently mind-blowing chips in my future smartphone, however.

Computer World has a write-up with quotes from the Intel engineers working on the multi-core smartphone chip, which is worth a read. What do you think about the prospects of a 48 core chip in your handheld mobile device?

AMD Announces It Will Build 64-bit ARM Processors for Server Markets

Subject: Processors | October 29, 2012 - 04:53 PM |
Tagged: processors, arm, amd, 64-bit

On a not very technically reliable webcast today, AMD has announced that it will produce 64-bit processors based on the ARM architecture and combine them with the "Freedom Fabric" they acquired with the purchase of SeaMicro.

amdarmlogos.jpg

In a move that is incredibly telling about the times we are in, but not really a surprise to those of us that follow the processor markets closely, AMD and ARM announced a partnership beyond previously discussed in public.  AMD will start production of ARM-based processors in 2014 and will be among the first to include 64-bit technology. 

The target for these processors will be the server market and AMD hopes to be at the forefront the often discussed ARM-in-the-server-world migration.  While that server opportunity size is debatable, with partners on stage like Facebook and RedHat, there is little doubt that it will have an affect on enterprise computing in the next 24 months.  AMD is hoping that its experience with the move to 64-bit technology in the x86 migration will aid them in development and migration in the ARM architecture world; one that is currently still limited to 32-bit. 

UPDATE: As being reported by Anand Shimpi this is in fact NOT an architecture license but is instead a processor license.  What does that mean?  AMD is not going to develop its own core (as Apple and NVIDIA do) but instead will fully integrate an upcoming 64-bit ARM core in new AMD products.

seamicro.jpg

Image source: EEBeat

SeaMicro's Freedom Fabric technology is another major angle that AMD has over other players in this field.  The fabric technology is meant to facilitate communication between multiple processors on a specialized bus, removing bottlenecks on the platform and network.  Dr. Lisa Su, SVP of Global Business at AMD, stated that simply connecting hundreds or thousands of ARM-based processors to each other isn't enough and moves the problem of computing management from the CPUs to the network itself.  Using Freedom Fabric, the AMD-based ARM processors would be able to much more efficiently communicate and thus maintain the promised power benefits of ARM servers.

AMD did state that they will continue to develop x86 processors going forward but you have to wonder about its dedication to that goal.  Working with ARM is a quick and easy way to get AMD into a growing market in the server world that Intel currently has no solutions for so it seems possible that this is simply a stop-gap until AMD can develop an x86-based solution.  It is hard to say for sure but for an organization in AMD's financial position, having options in multiple segments is certainly a good idea. 

seamicro2.jpg

What you won't see yet is AMD's graphics technology in the ARM-based processors announced today.  This isn't an "ARM APU" but instead is a combination of SeaMicro and ARM for a very specific server workload. 

We'll have more on this announcement if anything else interesting is divulged, but you can find the entire press release from AMD after the break!

Piledrivers are elegant in comparison to Bulldozers

Subject: Processors | October 23, 2012 - 02:44 PM |
Tagged: vishera, Steamroller, piledriver, FX-8350, fx-8150, FX-6300, FX-6200, bulldozer, amd

The FX-8350 Vishera processor from AMD has finally arrived with 8 fully unlocked cores of polished Piledriver processing power.  With Piledriver there are no huge changes to the existing Bulldozer architecture, this is more of a polishing and optimizing the existing architecture and [H]ard|OCP's testing bears that out.  While faster than the previous generation FX-8150 it still lags behind Intel's Ivy Bridge processors, disappointing but certainly expected.  The unlocked cores do lend themselves somewhat to overclocking, with [H] hitting a stable 4.6GHz with all cores enabled, a 10% jump in frequency.  At that speed it does better when competing with Intel's offerings, until you overclock them as well at which point the comparative performance suffers somewhat.

Make sure to catch Josh's review, covering both the 8 core FX-8350 and the $132 FX-6300 which has a disabled module; bringing back memories of older AMD chips whose modules could be brought back to life.

H_fx8350.png

"AMD's new Piledriver core technology should not be a surprise to any enthusiast as much of its "embargoed" information has already been exposed on the Net. Today we take the AMD FX series model 8350 desktop variant, code named Vishera, and look at it in an enthusiast way as we expose its IPC at 4GHz, and a bit of overclocking."

Here are some more Processor articles from around the web:

Processors

Source: [H]ard|OCP
Author:
Subject: Processors
Manufacturer: AMD

Bulldozer to Vishera

 

Bulldozer is the word.  Ok, perhaps it is not “the” word, but it is “a” word.  When AMD let that little codename slip some years back, AMD enthusiasts and tech journalists started to salivate about the possibilities.  Here was a unique and very new architecture that promised excellent single thread performance and outstanding multi-threaded performance all in a package that was easy to swallow and digest.  Probiotics for the PC.  Some could argue that the end product for Bulldozer and probiotics are the same, but I am not overly fond of writing articles containing four letter colorful metaphors.

vish_01.jpg

The long and short of Bulldozer is that it was a product that was pushed out too fast, it had specifications that were too aggressive for the time, and it never delivered on the promise of the architecture.  Logically there are some very good reasons behind the architecture, but implementing these ideas into a successful product is another story altogether.  The chip was never able to reach the GHz range it was supposed to and stay within reasonable TDP limits.  To get the chip out in a timely manner, timings had to be loosened internally so the chip could even run.  Performance per clock was pretty dismal, and the top end FX-8150 was only marginally faster than the previous top end Phenom II X6 1100T.  In some cases, the X6 was still faster and a more competent “all around” processor.

There really was not a whole lot for AMD to do about the situation.  It had to have a new product, and it just did not turn out as nicely as they had hoped.  The reasons for this are legion, but simply put AMD is competing with a company that is over ten times the size, with the resulting R&D budgets that such a size (and margins) can afford.  Engineers looking for work are a dime a dozen, and Intel can hire as many as they need.  So, instead of respinning Bulldozer ad nauseum and releasing new speed grades throughout the year by tweaking the process and metal layer design, AMD let the product line sit and stagnate at the top end for a year (though they did release higher TDP models based on the dual module FX-4000 and triple module FX-6000 series).  Engineers were pushed into more forward looking projects.  One of these is Vishera.

Click here to read the rest of the Vishera Review!

Hondo powered Win8 tablets; sweet spot or not?

Subject: Processors | October 18, 2012 - 05:58 PM |
Tagged: homdo, AMD z-series, z-60 apu, tablet, win8

Could AMD powered tablets firt in a sweet spot for those looking to pick up one of these new Win8 powered devices?  They will certainly be more powerful than an ARM powered WinRT tablet and the graphics will be superior to Intel powered tablets.  The Z-60 will have two 1GHz Bobcat cores each with 512KB of L2 cache and the HD 6250 GPU with 80 DirectX 11-class shader ALUs which should give snappy performance up to a 1920x1200 resolution.  The Tech Report talks about the various benefits and penalties to choosing a Hondo based device over an Ivy Bridge powered on in their article here.

amdhondoz60apu.jpg

"AMD is readying a new APU aimed at Window 8 tablets and hybrids. Otherwise known as Hondo, this Z-60 processor offers lower power consumption than AMD's existing APUs, and it comes with a side order of USB 3.0."

Here are some more Processor articles from around the web:

Processors

Intel Planning 10-core Xeon E5-2600 V2 Ivy Bridge-EP CPU

Subject: Processors | October 17, 2012 - 06:48 AM |
Tagged: xeon E5-2600 v2, lga 2011, Ivy Bridge-EP, Intel, 22nm

A recently leaked slide reveals one of Intel’s upcoming Xeon-branded server chips coming in Q3 2013. The Xeon E5-2600 V2 is an Ivy Bridge-EP processor and will be compatible with motherboards featuring the LGA 2011 socket.

Intel-Ivy-Bridge-EP-Launch-Roadmap.jpg

The Xeon E5-2600 V2 in particular has a 70W TDP (thermal design power) rating while the highest-end Ivy Bridge-EP CPUs will have TDPs of up to 130W. The E5-2600 V2 has 10 physical cores, and with HyperThreading it can handle a maximum of 20 threads. Each physical core has access to 256KB L2 cache and the chip has a total of 30MB L3 cache. Further, this (and other) Ivy Bridge-EP processor will support up to 1866MHz DDR3 system RAM.

Interestingly, the Xeon E5-2600 V2 is merely the middle of the road part for Intel. The company will be releasing processors that are even higher-end than this one. They will have up to 12 physical cores which means up to 24 threads. And paired with Intel's 22nm manufacturing process and 3D transistors, these chips will fit right into workstations and server rooms.

Source: Guru 3D

ASUS Maximus V Extreme-ly fast overclock.

Subject: General Tech, Motherboards, Processors | October 2, 2012 - 08:06 PM |
Tagged: overclock, asus

ASUSTeK has just accomplished a new world record overclock with their ASUS Maximus V Extreme motherboard. They calculated 1 million digits of Pi in a time of 5s 94ms which beats the current best time 5s 125ms according to HWBot. This result once validated lands the Maximus V Extreme in the 1st, 2nd, 3rd, and 4th place positions.

ASUS has once again broke records in the Pi eating contest with their Maximus V Extreme motherboard.

133595_10151433101032388_1117342732_o.jpg

It must be a fun day for an overclocker when you get to play with Liquid Helium. While I attended the Physics department of Queen’s University up here in Canada the facility was known for its condensed matter group. Much of the building was fitted with piping to recapture and recondense the Helium after its experiments strictly due to how much it cost and how rare it is. If someone offers for you to break an overclocking record with it you are obliged to say yes.

The achieved overclock appears to be tuned towards the application. Memory frequency was kept at 1333 MHz with a FSB of about 110 MHz. I would expect this multiplier-centric overclock is designed to keep the overclock focused on sheer number crunching which Super Pi likely relies on over memory bandwidth. Perhaps reduced memory timings might even come in to play for applications like this?

ASUS broke a few records with their Liquid Helium attempt. As of time of writing none of these records have been updated to the HWBot leaderboard.

1m.PNG

With Super Pi running to 1 million digits Asus and their team recorded a time of 5s 94ms -- 31 milliseconds faster than the current leading time of 5s 125ms. The current leaderboard already contains the ASUS Maximus V Extreme motherboard in Gold, Silver, and Bronze positions. This podium has already been well represented by the Maximus V.

When you cannot be satisfied with 1 million digits of pi you can run the marathon to 32 million digits.

superpi32M4min43.jpg

The most current record that I could find was set by a team sponsored by GSkill who achieved the time of 4min 44sec 609ms just a couple of weeks ago. ASUS and their team - which apparently has at least one member, “Smoke”, in common with the team GSkill assembled - also beat this record by almost 2 full seconds with a score of 4min 43s 0ms.

Finally, the real Trinity reviews arrive

Subject: Processors | October 2, 2012 - 04:56 PM |
Tagged: vishera, trinity, Steamroller, piledriver, bulldozer, amd, a8, a6, A4, a10, 5800K, 5600K

The NDA is over and we can finally tell you all about the new generation of Trinity, especially the compute portion which we were not allowed to discuss in the controversial preview.  Part of the good news is the price, Legit Reviews found the highest MSRP is $122 for the A10-5800K and it is currently available, though at $130.  The performance increase from the previous generation is decent for multicore applications though not so much for single threaded applications, overall you can expect general computing performance in line with Core i3 but not Core i5.  Gaming on the other hand did show much improvement, especially with you compare the built in HD7660D to Intel's current HD4000 and HD3500.  You can catch Josh's review right here.

LR_a10-apu-fm2.jpg

"The internal testing from AMD that we can see above shows a 37% increase in the 3DMark 11 score between the first generation A-Series Llano and this generation of A-Series Trinity. While our numbers don't match their numbers exactly, our Llano system scored 1115 3Dmarks while the AMD internal testing showed 1150 3DMarks. Our AMD A10-5800K scored 1521 3DMarks while they scored 1570. The overall difference was remarkably similar, AMD is boasting an increase of 37% and we saw a difference of 36.4%..."

Here are some more Processor articles from around the web:

Processors

Author:
Subject: Processors
Manufacturer: AMD

Trinity Finally Comes to the Desktop

Trinity.  Where to start?  I find myself asking that question, as the road to this release is somewhat tortuous.  Trinity, as a product code name, came around in early 2011.  The first working silicon was shown that Summer.  The first actual release of product was the mobile part in late Spring of this year.  Throughout the summer notebook designs based on Trinity started to trickle out.  Today we cover the release of the desktop versions of this product.

trin_01.jpg

AMD has certainly had its ups and downs when it comes to APU releases.  Their first real APU was Zacate, based on the new Bobcat CPU architecture.  This product was an unmitigated success for AMD.  Llano, on the other hand, had a pretty rocky start.  Production and various supply issues caused it to be far less of a success than hoped.  These issues were oddly enough not cleared up until late Spring of this year.  By then mobile Trinity was out and people were looking towards the desktop version of the chip.  AMD saw the situation, and the massive supply of Llano chips that it had, and decided to delay introduction of desktop Trinity until a later date.

To say that expectations for Trinity are high is an understatement.  AMD has been on the ropes for quite a few years in terms of CPU performance.  While the Phenom II series were at least competitive with the Core 2 Duo and Quad chips, they did not match up well against the latest i7/i5/i3 series of parts.  Bulldozer was supposed to erase the processor advantage Intel had, but it came out of the oven as a seemingly half baked part.  Piledriver was designed to succeed Bulldozer, and is supposed to shore up the architecture to make it more competitive.  Piledriver is the basis of Trinity.  Piledriver does sport significant improvements in clockspeed, power consumption, and IPC (instructions per clock).  People are hopeful that Trinity would be able to match the performance of current Ivy Bridge processors from Intel, or at least get close.

So does it match Intel?  In ways, I suppose.  How much better is it than Bulldozer?  That particular answer is actually a bit surprising.  Is it really that much of a step above Llano?  Yet another somewhat surprising answer for that particular question.  Make no mistake, Trinity for desktop is a major launch for AMD, and their continued existence as a CPU manufacturer depends heavily on this part.

Continue reading our review of the AMD Trinity A10 APUs!!