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

Operating System U Fails To Live Up To Its Goals

After launching last month on Kickstarter, the project has turned into a failure and all development has ceased. Operating System U by Andrew Bernstein only raised $1,948 of its $50,000 goal over the month-long period for the OS that claimed numerous advantages over Ubuntu and Windows 8. Andrew then posted, "Unfortunately OS U was unsuccessful. I truly, truly appreciate everyone who backed us, but unfortunately since we where unsuccessful, combined with other circumstances, OS U will not have any more continued development." Read more

Calculate Intro, OpenMandriva Review, and Mageia Delay

Today in Linux news Jessie Smith has a nice article on Gentoo-derivative Calculate Linux 14 in this week's Distrowatch Weekly. Linuxbsdos.com has a review of OpenMandriva Lx 2014.1, released last week. Mageia 5 Beta 1 is delayed and openSUSE 11.4 is "truly, finally dead." We have all this and more in tonight's Linux news recap. Read more

Early Morning Linux Voodoo at Denny’s

I could tell that he wasn’t comfortable turning over control of his laptop to a stranger, but after a few seconds I got a slight nod to the affirmative. I pulled the Acer over to my part of the counter and booted the Linux Mint KDE LTS I keep for just such purposes. As the computer accepted the DataStick as the boot option, I explained to Ed what I was doing. It was obvious he had no idea what I was talking about so we waited in awkward silence for the next few seconds. Finally, the Mint logo appeared on the screen. I opened Dolphin and located the Windows drive then asked him for the name of the file. He couldn’t remember but was sure it was a PDF. A few minutes later, I pulled a pen from my pocket and wrote down the number he needed and slid it back over to him with his laptop. Read more

Leftovers: Proprietary Software and Command Line