Language Selection

English French German Italian Portuguese Spanish

Nvidia scientist calls for expanded research into parallelism

Filed under
Hardware

Expanded research is needed into techniques for identifying and preserving parallelism in chip applications, according to chip maker Nvidia's chief scientist.

Comparing the futures of general-purpose CPUs and graphics processors (GPUs), David Kirk told the 17th Hot Chips conference on the Stanford Univeristy campus here on Tuesday (Aug. 16) that a crisis looms in programming technology. He said this could not only blight the future CPU market but also bring an end to improvements in graphics performance despite continued improvements in GPUs.

"If we look at the situation of general-purpose CPUs," Kirk said, "we see architects moving rapidly to multithreading and multicore designs. But we don't see a lot more threads to run. What parallelism there may be in algorithms is often concealed or lost altogether in the programming process."

Kirk grounded his pessimism in the experiences of game developers trying to exploit new multicore CPU chips. "We are already seeing some games limited by CPU throughput. We can render images faster than the CPU can send us the data," Kirk said. "But when game developers try to use dual-core CPUs to help, we have seen virtually no benefit to the second CPU core. And if the developer doesn't clearly understand the interactions of the cores with the caches, we have seen the application actually run slower on a dual-core machine."

Kirk contrasted this situation against the entirely different structure inside the GPU. "Graphics has been called embarrassingly parallel," he said. "In effect, each stage in our pipeline, each vertex in the scene and each pixel in the image is independent. And we have put a lot of effort into not concealing this parallelism with programming."

This allows a GPU developer to simply add more vertex processors and shading engines to handle more vertices and more pixels in parallel, as process technology allows. "We are limited by chip area, not by parallelism," Kirk observed.

Full Story.

More in Tux Machines

Kernel Backports and Graphics

  • [Older] Backports and long-term stable kernels
  • What’s New in Wayland and Weston 1.12?
    The Wayland core protocol documentation has received numerous refinements to improve its clarity and consistency. Along with this, many blank areas of the protocol documentation have been fleshed out. A new wl_display_add_protocol logger API provides a new, interactive way to debug requests; along with this are new APIs for examining clients and their resources. This is analogous to using WAYLAND_DEBUG=1, but more powerful since it allows run time review of log data such as through a UI view. There have been improvements to how the protocol XML scanner handles version identification in protocol headers. This enables better detection and fallback handling when compositors and clients support differt versions of their protocols.
  • XDC2016 Wraps Up After Many Wayland, X.Org & Mesa Discussions
    The 2016 X.Org Developers' Conference (XDC2016) wrapped up Friday in Helsinki, Finland. Here is a summary of the major happenings for those that may have missed it or didn't yet watch the video streams.

IBM Claims “New Linux Based Power System Server Kicks Butt

today's howtos

Leftovers: Ubuntu

  • Ubuntu Phone, Sep 2016 - Vorsprung durch Touch
    The Ubuntu Phone is getting better, and with every new iteration of the OTA, my little BQ Aquaris E4.5 is gaining more speed and functionality. Like in the air force, with an avionics upgrade, which transforms ancient wings into a powerful and modern bird of prey. Only the pace of advancement is lagging behind the market. See what Android and iOS can do, even Windows Phone, and you realize how late and insufficiently meaningful the Ubuntu Phone really is. This has to change, massively. This latest round does bring some fine goods to the table - more speed and stability, better icons, more overall visual polish, incremental improvements in the applications and the scopes. But that's not enough to win the heart of the average user. A more radical, app-centric effort is required. More focus on delivering the mobile experience, be it as it may. Ubuntu cannot revolutionalize that which is already considered the past. It can only join the club and enjoy the benefits of a well-established reality. And that is a kickass app stack that makes the touch device worth using in the first place. Still, it's not all gloomy. E4.5 is a better product now than it was a year ago, fact. Ubuntu Phone is a better operating system than it was even this spring, fact. So maybe one day we will see Ubuntu become an important if not dominant player in the phone and tablet space. It sure is heading in the right direction, my only fear is the availability of resources to pull off this massive rehaul that is needed to make it stand up to the old and proven giants. And that's it really. If you're keen on Linux (not Android) making it in the mobile world, do not forget to check my Ubuntu tablet review! Especially the convergence piece. On that merry note, you do remember that I'm running a wicked contest this year, too? He/she who reads my books might get a chance to win an M10 tablet. Indeed. Off you go, dear readers. Whereas I will now run the same set of tests we did here on the Aquaris tablet, and see how it likes the OTA-12 upgrade. The end.
  • Ubuntu 16.10 Unity 8 - new window snapping feature
  • Ubuntu Online Summit for Ubuntu 17.04 is Taking Place In Mid-November
  • Ubuntu Online Summit: 15-16 November 2016