AMD Quietly Rolls Out hUMA – Potential Game-Changer for Parallel Computing

Background  High Performance Attached Processors Handicapped By Architecture

The application of high-performance accelerators, notably GPUs, GPGPUs (APUs in AMD terminology) to a variety of computing problems has blossomed over the last decade, resulting in ever more affordable compute power for both horizon and mundane problems, along with growing revenue streams for a growing industry ecosystem. Adding heat to an already active mix, Intel’s Xeon Phi accelerators, the most recent addition to the GPU ecosystem, have the potential to speed adoption even further due to hoped-for synergies generated by the immense universe of x86 code that could potentially run on the Xeon Phi cores.

However, despite any potential synergies, GPUs (I will use this term generically to refer to all forms of these attached accelerators as they currently exist in the market) suffer from a fundamental architectural problem — they are very distant, in terms of latency, from the main scalar system memory and are not part of the coherent memory domain. This in turn has major impacts on performance, cost, design of the GPUs, and the structure of the algorithms:

  • Performance — The latency for memory accesses generally dictated by PCIe latencies, which while much improved over previous generations, are a factor of 100 or more longer than latency from coherent cache or local scalar CPU memory. While clever design and programming, such as overlapping and buffering multiple transfers can hide the latency in a series of transfers, it is difficult to hide the latency for an initial block of data. Even AMD’s integrated APUs, in which the GPU elements are on a common die, do not share a common memory space, and explicit transfers are made in and out of the APU memory.
Read more

No Matter The Mobile Technology, Extend Your Product Packaging To Engage Customers

My colleague Reineke Reitsma recently published a blog on the limited but growing uptake of QR/2D barcodes.

Let’s face reality. Usage is low and marketing execution is poor to date, with too many campaigns that lack a clear consumer benefit and that provide a bad user experience by not offering mobile-optimized content. Today, mobile bar codes are an interesting tactic to engage with early adopters.

However, moving forward, we expect QR codes to gain traction and to be increasingly mixed with other technologies (including radio technologies like NFC) to provide extended product packaging solutions. Bar codes do not have to be just cold, emotionless, black-and-white squares. Solutions now exist to personalize QR codes’ designs and seamlessly mix them into a logo or band chart – even merging QR codes and NFC tags, as in the example below from mobiLead solutions.

The 2D bar code market will follow the same path as the 1D bar code market: fulfilling the need for certified and scalable platforms dealing with millions of standard code generation. Mobile bar code vendors will have to move into scalable mobile engagement platforms, progressively integrating multiple access technologies, such as Near Field Communications (NFC) tags, image recognition, or audio tags such as Shazam, and offering deep analytical tools. Beyond the emerging role of 2D bar codes in sales, we expect a growing number of brands — especially in the nutrition and health space — to systematize the use of bar codes on product packaging. Consumers want access to more product information, and brands can leverage mobile technologies to create a consumer relationship.

Read more