AMD And Oracle Announce GPU Accelerated Java: An Investment Opportunity

Throughout the years, there have been a number of critical advancements in CPU technology that have fundamentally changed the landscape. Integrated floating point co-processors, on-die caches and memory controllers, and multiple processor cores on a single chip are just a few highlights of the innovations that have enabled fundamentally new levels of performance and efficiency.

Advanced Micro Devices (NASDAQ:AMD) and Oracle (ORCL) have just announced that the former will be working with the latter (and a number of other partners) to allow Java programs to seamlessly take advantage of the graphics processing unit to very significantly improve performance in multi/many-core environments.

Now, to understand the importance of this announcement, it is important that I explain two key technologies: Java and general purpose GPU or "GPGPU".

Java - The Language Of Modern Software

Java is a programming language. The reason that this programming language is so special is that it is both widely used and programs written in it will run on any platforms that support Java. Unlike "native" languages such as C/C++, Java code is translated into an "intermediate" language, called "bytecode." This is then fed into a program called the "Java Virtual Machine" that translates this platform-agnostic bytecode into instructions that the particular platform can understand. As long as a particular instruction set architecture (x86, ARM (ARMH), MIPS (MIPS), etc.) has a Java Virtual Machine written for it, the Java programs will be compatible.

GPGPU - A Way To Increase Compute Efficiency

The next important technology is the graphics processing unit applied to general purpose tasks - GPGPU. When graphics processors first hit, they were useful for pretty much one thing: drawing 3D graphics at a much faster rate than CPUs due to dedicated hardware.

Eventually, though, graphics processing demanded more flexibility in the operations that were needed. This led to the graphics processing unit evolving into a highly programmable, compute-density focused piece of silicon. Per watt, and per unit of silicon die area, graphics processing units provide orders of magnitude more brute-force processing power than a traditional CPU. So, why not get rid of the CPU altogether?

The catch is that GPUs get their raw processing power from having a whole bunch of fairly specialized compute units - they're good at just doing a a lot of raw calculations. A good deal of a traditional general purpose CPU is dedicated to things like branch predictors (units that try to guess what path a program's execution will take), caches, prefetchers, and other things that promote fast single-threaded performance.

So, the "big idea," then, is to take full advantage of the integration of the CPU and GPU in modern processors and leverage that raw power in general purpose applications that can take advantage of it. So, what are some programs that are conducive to GPGPU acceleration?

• Video processing
• Audio processing
• Computational finance
• Weather forecasting
• Medical imaging
• Electronic design automation

This is not even close to an exhaustive list; for a more complete reference, I recommend this excellent page.

Bringing GPGPU To The Masses

Interestingly enough, GPGPU has seen fairly wide adoption. Both Nvidia (NVDA) and AMD offer discrete accelerator boards that are used for these applications (these are generally the standard consumer graphics boards with features enabled and custom firmware).

More importantly, the standard consumer graphics products from the aforementioned vendors also support GPGPU processing, but these are generally less feature filled than the dedicated products.

However, a fairly small portion of the population will buy a $3,000 dedicated HPC accelerator, or even a $200+ graphics board. So, the question is: how can this be brought to the masses?

See, the GPGPU hardware needs to be in the hands of the masses before developers will seriously consider making the investment to get on board. This is what AMD's strategy is with Heterogeneous System Architecture or "HSA." Every mainstream client processor that AMD ships has built in graphics that can run GPGPU code. Further, as Intel (INTC) has started to make a more serious attempt to bundle its CPUs with capable GPUs, it will also help provide the hardware install base.

It's Up To The Software Companies

The seeds are now sown on the hardware side, so it is up to the software developers to make it happen. This is why the recent announcement that Oracle will build support into its Java platform for GPGPU acceleration in multi-core environments is huge. The easier and more transparent it is for programmers to leverage this technology, the more rapidly it will see adoption.

The Investment Opportunity

With sufficient background established, I will now point out that GPGPU is a very significant technology and that the leaders in this space will very quickly become the more relevant players in high performance, power-efficient computing. The efficiency and raw power of GPGPU will be important in both client and server segments, and with major software support from Oracle, Mathworks, and others, the technology will eventually become critical across the entire compute continuum.

So, right now, the two best bets in the space are, naturally, the GPU vendors, AMD and Nvidia. AMD has a lead in integrating its GPGPU technology into low cost, mainstream parts and in deepening that integration with each iteration of its CPUs. Nvidia's software support with its proprietary "CUDA" platform has been excellent, however, which is why the company's accelerators are found in a number of Top 500 supercomputers.

So AMD's has the client/mainstream whereas Nvidia is winning over the hearts of the HPC/supercomputer crowd with dedicated accelerators. I believe that both will prosper, but AMD's presence in the PC/notebook/server spaces gives it a much larger total addressable market.

The real question is how will this affect the Intel/AMD dynamic going forward? Intel's CPU designs are typically faster and more power efficient, but AMD's GPUs are faster and more feature-filled than Intel's. In a world where GPGPU becomes a major performance driver, AMD would become fiercely competitive, which could drive its ASPs (and gross margin profile) up considerably in all its major segments.

This first step to GPGPU for the masses via Oracle and Java is the key to the realization of AMD's "heterogeneous system architecture" plan. With the software support in place, AMD's products become a lot more compelling and competitive - improving sales, margins, and ultimately profitability.

Disclosure: I am long AMD, INTC, NVDA. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Comments

[Sal Marvasti]

Interesting take, though more important is current tech that is available now. I.e. C++ AMP, OpenCL. I used to work for mathworks , it was the group I was in that added GPGPU. I believe at the moment it is CUDA based so NVDA only. But they plan to switch to OpenCl.

[jefbe80]

ashraf thanks for sharing your knowledge it's enriching and useful, and always it'ss good to read your articles

[Ashraf Eassa]

Thank you for the kind words!

[Whitehawk]

I found these pages helpful:
http://bit.ly/VXOzGZ
http://bit.ly/T2Hlge

AMD bought its way into this space years ago with the ATI Tech acquisition. Have they kept up with NVDA?

BTW: AMD sold part of its GPU IP over the years, namely Imageon to Qcom. This is a different space than GPGPU, but the question is: Good decision?

[IAmNetburst]

I don't see how the pages are helpful, except to graphics card technicians. How many SA readers know how to compare unified shaders, texture shaders, ROPs, memory bus, etc.? More likely, they will simply look at clock rates and think they can learn something about performance.

Nvidia does have far better marketing, though. I've never heard of the IP sale to Qcom referred to as a good decision...

[anand]

Very respectfully IAmNetburst I have carefully studied these two wiki pages for Nvidia and ATI GPUs many years. Note that they are not always current. ;-)

Whitehawk AMD is behind Nvidia on GPGPUs. AMD is behind the industry in servers.

Obviously AMD would rather not have sold Imageon, but they could have had a near liquidity event a couple years ago.

AMDFANBOY, AMD's tablet chip isn't quite good enough yet. AMD still is not very good at the whole SoC package. I don't think AMD be able to introduce a highly competitive tablet chip until 20 nm skews two years from now. Assuming AMD hits its benchmarks.

Ashraf, I still lack enough information to comment intelligently on the Oracle announcement:

"Oracle disclosed that it will be using its HotSpot JVM and the libraries from Java 8's Lambda project, which was published last year as a way to support Java programming in multicore environments. If a GPU is available in a system, Java code will be converted to OpenCL code and then run on the GPU. In a statement released to the press, Georges Saab, vice president of software development for the Java Platform Group at Oracle said: "We expect our work with AMD and other OpenJDK participants in Project 'Sumatra' will eventually help provide Java developers with the ability to quickly leverage GPU acceleration for better performance.""

I have seen other similar announcements before about how software will convert to OpenCL code and then run on a GPU. Let us see how efficient HotSpot JVM and the libraries from Java 8's Lambda project are at converting Java code to OpenCL and efficiently running tasks. Easier said than done.

You might contact the Anandtech guys for comment.

This could simply mean that Oracle is trying to get Java [and potentially software such as database and enterprise software] to run faster on X86 based servers, desktops and notebooks. Which is of course what Oracle should do to compete with rival software providers and rival Software as a Service cloud platforms.

Could this also presage that Oracle wants AMD to develop accelerator co-processors [FirePro successors] that are designed to work with Sparc servers? If so, then this potentially gets interesting. Is Oracle serious about taking on Xeon and Xeon Phi with Sparc plus AMD/ATI accelerator co-processors [and ideally later persuading Nvidia to similarly design workstation Tesla GPUs that work well with Sparc]?

Please share any new information you find out. I will do some digging of my own. There is a lot more research to do.

[BTM]

This is definitely not a big deal. Most users don't need more processing power. CUDA and Open CL have been around now for years and have only had a minor impact. I don't see this as being any different.

[AMDFANBOY]

It might become a deal after 1-2 years if Oracle decides to build APU-servers with AMD.

But even if not I would like to discuss a different issue: Tablets.

Many analysts say that the PC is going to die more or less because of ARM and the Tablets.

However, after Intel AMD just released an own tablet chip.

I see a large business opportunity here. Just stick a keyboard for 5 bucks into the usb-slot and a mouse and you have a pc.

No Android tablet can offer this and even Apple looks behind here because you cannot turn an iPad into a Mac!

This big advantage could change the market in favour of Wintel and AMD in a few years!

[AMDFANBOY]

This here is only one Lego part in the game. I think in the future we will see more GPU power and less that of the CPU because a GPU has much more computing power at very low energy consumption.

We must also consider that both ARM/RISC and x86 include almost the same kind of controller functionality for i/o as well as cache and so on with the core becoming less important.

Therefore, the power saving advantages of ARM and RISC in general will disappear because the main force will be the GPU, while the CPU is more important for compatibility.

Here is the true strength of AMD: very strong GPU power for heavy workloads combined with the x86 compatibility of an also quite competitive PC-Processor.

No other company in the world can provide this.

[rcketscientist]

It's not really a big deal. Any project large enough to want to leverage GPGPU, that for some reason chose java, would already have a native component in their software for a variety of reasons. How does AMD have the lead just hopping into the scene, when nVidia has been the mainstream pioneer and the status quo on the technology?