Sign up for our Technology Newsletter 
nVidia Releases CUDA
Nvidia has released CUDA – its code that lets developers run their code on GPUs – to server vendors in order to get 64-bit ARM cores into the high performance computing (HPC) market.
The firm said today that ARM64 server processors, which are designed for microservers and web servers because of their energy efficiency, can now process HPC workloads when paired with GPU accelerators using the Nvidia CUDA 6.5 parallel programming framework, which supports 64-bit ARM processors.
“Nvidia’s GPUs provide ARM64 server vendors with the muscle to tackle HPC workloads, enabling them to build high-performance systems that maximise the ARM architecture’s power efficiency and system configurability,” the firm said.
The first GPU-accelerated ARM64 software development servers will be available in July from Cirrascale and E4 Computer Engineering, with production systems expected to ship later this year. The Eurotech Group also plans to ship production systems later this year.
Cirrascale’s system will be the RM1905D, a high density two-in-one 1U server with two Tesla K20 GPU accelerators, which the firm claims provides high performance and low total cost of ownership for private cloud, public cloud, HPC and enterprise applications.
E4′s EK003 is a production-ready, low-power 3U dual-motherboard server appliance with two Tesla K20 GPU accelerators designed for seismic, signal and image processing, video analytics, track analysis, web applications and Mapreduce processing.
Eurotech’s system is an “ultra-high density”, energy efficient and modular Aurora HPC server configuration, based on proprietary Brick Technology and featuring direct hot liquid cooling.
Featuring Applied Micro X-Gene ARM64 CPUs and Nvidia Tesla K20 GPU accelerators, the new ARM64 servers will provide customers with an expanded range of efficient, high-performance computing options to drive compute-intensive HPC and enterprise data centre workloads, Nvidia said.
Nvidia added, “Users will immediately be able to take advantage of hundreds of existing CUDA-accelerated scientific and engineering HPC applications by simply recompiling them to ARM64 systems.”
ARM said that it is working with Nvidia to “explore how we can unite GPU acceleration with novel technologies” and drive “new levels of scientific discovery and innovation”.
nVidia Outs CUDA 6
Nvidia has made the latest GPU programming language CUDA 6 Release Candidate available for developers to download for free.
The release arrives with several new features and improvements to make parallel programming “better, faster and easier” for developers creating next generation scientific, engineering, enterprise and other applications.
Nvidia has aggressively promoted its CUDA programming language as a way for developers to exploit the floating point performance of its GPUs. Available now, the CUDA 6 Release Candidate brings a major new update in unified memory access, which lets CUDA applications access CPU and GPU memory without the need to manually copy data from one to the other.
“This is a major time saver that simplifies the programming process, and makes it easier for programmers to add GPU acceleration in a wider range of applications,” Nvidia said in a blog post on Thursday.
There’s also the addition of “drop-in libraries”, which Nvidia said will accelerate applications by up to eight times.
“The new drop-in libraries can automatically accelerate your BLAS and FFTW calculations by simply replacing the existing CPU-only BLAS or FFTW library with the new, GPU-accelerated equivalent,” the chip designer added.
Multi-GPU Scaling has also been added to the CUDA 6 programming language, introducing re-designed BLAS and FFT GPU libraries that automatically scale performance across up to eight GPUs in a single node. Nvidia said this provides over nine teraflops of double-precision performance per node, supporting larger workloads of up to 512GB in size, more than it’s supported before.
“In addition to the new features, the CUDA 6 platform offers a full suite of programming tools, GPU-accelerated math libraries, documentation and programming guides,” Nvidia said.
The previous CUDA 5.5 Release Candidate was issued last June, and added support for ARM based processors.
Aside from ARM support, Nvidia also improved Hyper-Q support in CUDA 5.5, which allowed developers to use MPI workload prioritisation. The firm also touted improved performance analysis and improved performance for cross-compilation on x86 processors.
nVidia Pays Up
January 10, 2014 by admin
Filed under Around The Net
Comments Off on nVidia Pays Up
Nvidia has agreed to pay any Canadian who had the misfortune to buy a certain laptop computer made by Apple, Compaq, Dell, HP, or Sony between November 2005 and February 2010. Apparently these models contained a dodgy graphics card which was not fixed for five years.
Under a settlement approved by the court Nvidia will pay $1,900,000 into a fund for anyone who might have bought a faulty card. The Settlement Agreement provides partial cash reimbursement of the purchase price and you have to submit a claim by February 25, 2014. You will know if your Nvidia card was faulty because your machine would have a distorted or scrambled video, or no video on the screen even when the computer is on. There would be random characters, lines or garbled images – a bit like watching one of the Twilight series. There will be intermittent video issues or a failure to detect wireless adaptor or wireless networks.
The amount of compensation will be determined by the Claims Administrator who will apply a compensation grid and settlement administration guidelines. Cash compensation will also be provided for total loss of use based on the age of the computer; temporary loss of use having regard to the nature and duration of the loss of use; and reimbursement for out-of-pocket expenses caused by Qualifying Symptoms to an Affected Computer.
nVidia Launching New Cards
We weren’t expecting this and it is just a rumour, but reports are emerging that Nvidia is readying two new cards for the winter season. AMD of course is launching new cards four weeks from now, so it is possible that Nvidia would try to counter it.
The big question is with what?
VideoCardz claims one of the cards is an Ultra, possibly the GTX Titan Ultra, while the second one is a dual-GPU job, the Geforce GTX 790. The Ultra is supposedly GK110 based, but it has 2880 unlocked CUDA cores, which is a bit more than the 2688 on the Titan.
The GTX 790 is said to feature two GK110 GPUs, but Nvidia will probably have to clip their wings to get a reasonable TDP.
We’re not entirely sure this is legit. It is plausible, but that doesn’t make it true. It would be good for Nvidia’s image, especially if the revamped GK110 products manage to steal the performance crown from AMD’s new Radeons. However, with such specs, they would end up quite pricey and Nvidia wouldn’t sell that many of them – most enthusiasts would probably be better off waiting for Maxwell.
nVidia’s CUDA 5.5 Available
Nvidia has made its CUDA 5.5 release candidate supporting ARM based processors available for download.
Nvidia has been aggressively pushing its CUDA programming language as a way for developers to exploit the floating point performance of its GPUs. Now the firm has announced the availability of a CUDA 5.5 release candidate, the first version of the language that supports ARM based processors.
Aside from ARM support, Nvidia has improved supported Hyper-Q support and now allows developers to have MPI workload prioritisation. The firm also touted improved performance analysis and improved performance for cross-compilation on x86 processors.
Ian Buck, GM of GPU Computing Software at Nvidia said, “Since developers started using CUDA in 2006, successive generations of better, exponentially faster CUDA GPUs have dramatically boosted the performance of applications on x86-based systems. With support for ARM, the new CUDA release gives developers tremendous flexibility to quickly and easily add GPU acceleration to applications on the broadest range of next-generation HPC platforms.”
Nvidia’s support for ARM processors in CUDA 5.5 is an indication that it will release CUDA enabled Tegra processors in the near future. However outside of the firm’s own Tegra processors, CUDA support is largely useless, as almost all other chip designers have chosen OpenCL as the programming language for their GPUs.
Nvidia did not say when it will release CUDA 5.5, but in the meantime the firm’s release candidate supports Windows, Mac OS X and just about every major Linux distribution.
Are CUDA Applications Limited?
Acceleware said at Nvidia’s GPU Technology Conference (GTC) today that most algorithms that run on GPGPUs are bound by GPU memory size.
Acceleware is partly funded by Nvidia to provide developer training for CUDA to help sell the language to those that are used to traditional C and C++ programming. The firm said that most CUDA algorithms are now limited by GPU local memory size rather than GPU computational performance.
Both AMD and Nvidia provide general purpose GPU (GPGPU) accelerator parts that provide significantly faster computational processing than traditional CPUs, however they have only between 6GB and 8GB of local memory that constrains the size of the dataset the GPU can process. While developers can push more data from system main memory, the latency cost negates the raw performance benefit of the GPU.
Kelly Goss, training program manager at Acceleware, said that “most algorithms are memory bound rather than GPU bound” and “maximising memory usage is key” to optimising GPGPU performance.
She further said that developers need to understand and take advantage of the memory hierarchy of Nvidia’s Kepler GPU and look at ways of reducing the number of memory accesses for every line of GPU computing.
The point Goss was making is that GPU computing is relatively cheap in terms of clock cycles relative to the time it takes to fetch data from local memory, let alone loading GPU memory from system main memory.
Goss, talking to a room full of developers, proceeded to outline some of the performance characteristics of the memory hierarchy in Nvidia’s Kepler GPU architecture, showing the level of detail that CUDA programmers need to pay attention to if they want to extract the full performance potential from Nvidia’s GPGPU computing architecture.
Given Goss’s observation that algorithms running on Nvidia’s GPGPUs are often constrained by local memory size rather than by the GPU itself, the firm might want to look at simplifying the tiers of memory involved and increasing the amount of GPU local memory so that CUDA software developers can process larger datasets.
Will Tegra 4 Launch In Q2?
Tegra 4 was supposed to be production ready in Q4 2012 and the general expectation was that CES 2013 would be marked by the launch of phones and tablets based on the new chipset.
It turns out that the chip needed another re-spin, something that usually creates a delay of roughly a quarter. We don’t know which part of the chip was to blame but our sources claim that Tegra 4 is a complex chip with a lot of components where many things can go wrong.
Nvidia dared to move to 28nm, change the core from A9 to A15 and find a way to make its LTE work. There were a lot of things that could go wrong and obviously some did.
This is why Intel first shrinks the core, for example from 32nm to 22nm, and then in its “tock” cycle goes for a newly designed core. Nvidia doesn’t have that luxury, as making a 28nm version of Tegra 3 would not be enough for the SoC market in 2013.
A few people at Nvidia have been telling us that the chip has been sampled to accounts and Nvidia is planning to have some designs announced at the Mobile World Congress. We managed to confirm this schedule with some Nvidia partners.
nVidia’s Tegra 4 Specs Spotted
Here is an interesting leak, just what the doctor ordered to spice up a rather slow news cycle. Chiphell has posted a slide containing a few Tegra 4 specs, but we still don’t know the clocks or a few other interesting details. Of course, the leak should be taken with a grain of salt, but the specs are more or less in line with what we were expecting all along.
Tegra 4, codenamed Wayne, is a 28nm part with revamped graphics and new ARM cores. Although the slide does not directly point to the type of ARM cores used in the design, the new chip is based on ARM’s latest A15 core. Like the Tegra 3, the new chip will also feature an additional companion core to improve energy efficiency. No surprises here really.
In terms of GPU performance, Nvidia promises to deliver a six-fold improvement over the Tegra 3 and a 20x improvement over Tegra 2 chips. Oddly enough, in spite of Nvidia’s graphics prowess, Tegra chips never featured world-beating graphics. This time around they could, thanks to the new 72-core GPU. The GPU will be able to cope with 2560×1600 screens at 120Hz, but it could also take on 4K resolutions, although details are still sketchy. At this point 4K support could only be relevant for next-generation smart TVs, with a huge price tag.
As far as other features go, Tegra 4 brings support for USB 3.0 and DDR3L dual-channel memory. The leak does not mention LTE support.
Tegra 4 will have to take on the likes of Samsung’s upcoming Exynos 5440, which should also debut in early 2013. Nvidia was first to market with a quad-core A9 chip, but this time around it will have to face off against the new Exynos and A15 quad-cores from other vendors.
Nvidia is expected to showcase the new chip at CES and we’ll be there to check it out.
nVidia Soars
Nvidia has published its third quarter earnings and the results are impressive to say the least. With record revenue of $1.2 billion, Nvidia’s net income in Q3 was $209.1 million (GAAP).
Quarterly revenue is up 12.9 percent year-over-year and represents a 15.3 percent sequential bump, beating analyst expectations. However, Nvidia expects its revenue to dip to between $1.02 and $1.17 billion in the fourth quarter.
The company blames the projected slump on a declining PC market. It seems Nvidia does not expect Windows 8 will have a very positive impact on the PC market.
Most Tegra 2 Tablets Will Get ICS
Sources have confirmed that most Tegra 2 tablets you know will get Ice Cream Sandwich. We are still sniffing around to find out if the ICS is going to end up as Android 4.0 but it will bring phones and tablets much closer and should ship in October or November.
Many Asus, Samsung, Toshiba, Lenovo, Sony and any other Android 3.x compatible tablets on market will have a chance to get the new one. The upgrade will come as manufacturers get it ready and customized for its tablets but most tablets will ship with Android 4.0, Ice Cream Sandwich, probably early next year at the latest.
This is good news for many who were brave to buy the first generation of tablets not based on Apple’s architecture and it will help Google to gather even more momentum for 2012. 2012 looks like a year when Google will be ready for real war against Apple, but at the same time, Android supporters fear that Windows 8 will get a lot of attention when it ships in late 2012.