Downsizing Multicore Programming Skills to Take Advantage of Intel Atom

Posted by Gaston Hillar @ 02:55:PM | October 2, 2009

There is no doubt that Intel Atom is an attractive microprocessor. Netbooks, embedded systems, consumer electronic devices, and MIDs (Mobile Internet Devices) are using its different flavors. But can multicore programming skills be useful with these single-core CPUs?

I spent last week attending Intel Developer Forum 2009 where the most popular words were "netbooks," "Atom," "multicore," and "parallelism." I had a great interest on the new Nehalem EX with its 8 physical cores and 16 hardware threads (16 logical cores when Hyper-Threading is enabled). Nonetheless, I was also involved in finishing an application to take advantage of the power offered by a tiny Intel Atom microprocessor running on netbooks and on MIDs (Mobile Internet Devices).

Intel Atom is a very powerful microprocessor. It is a single-core CPU. However, it offers Hyper-Threading, therefore you have ttwo hardware threads, with many shared resources. I wanted to show an application using nice visual effects, displaying images with animations and extra large icons. I had read Stephen Blair-chappell's article Parallel Execution Advantage on Single Core Intel Atom. His experience with an Intel Atom netbook motivated me to design an application with parallelism in mind, trying to take advantage of two hardware threads. It was a really difficult task because I had to target Windows XP, Windows Vista, Windows 7, Moblin 2.1 and Midinux 3.0.

When you target multiple platforms, you usually lose some performance, there is always a trade-off. However, if you design an application with parallelism in mind, you will gain performance in most modern microprocessors, even on Intel Atom. I could improve the overall performance by 35% taking advantage of Hyper-Threading. However, it was not enough to create real-time animations using multiple effects. As Intel Atom also offers MMX; SSE; SSE2; SSE3 and SSSE3, I had the possibility to take advantage of these powerful additional instructions. As I was working with visual effects, I could improve the overall performance by 80%.

CPU-Z displaying the features offered by Intel Atom N270 (MMX; SSE; SSE2; SSE3; SSSE3 and 2 hardware threads).

Finally, I had the possibility to run the application on stage as part of the keynote launching the Intel Atom Developer Program.

The good news is that you can downsize your multicore and parallel programming experience to target new mobile microprocessors and you'll be able to achieve very interesting performance improvements. Besides, I believe we will see multicore Atom CPUs in the future.

No matter whether you're targeting an eight-core CPU or a tiny single-core CPU with Hyper-Threading, you can go parallel in order to take advantage of the newest micro-architectures.

This Week's Multicore Reading List
MATLAB and Google App Engine
Logging In C++ : Part 2
 Improving log granularity
A Conversation with BitMagic's Developer
Prefer Structured Lifetimes: Local, Nested, Bounded, Deterministic

All Feature Stories

Real World Parallelism Webinar Series

  • November 17, 2009
    Visual Effects for Animation - presented by DreamWorks Animation
    Speaker: Ron Henderson (Bio)

    Ron Henderson manages the FX Tools group at DreamWorks Animation, where he is responsible for developing physical simulation and procedural modeling tools. These systems have been used for key visual effects in recent films such as Kung Fu Panda and Monsters vs. Aliens (March 2009).

    Prior to joining DreamWorks in 2002 he was a senior scientist at Caltech with a joint appointment to the Applied Math and Aeronautics departments, where he worked on efficient techniques for the direct numerical simulation of fluid turbulence.

    Abstract:
    In this webinar, Ron Henderson will show examples of visual effects, from hair and feathers to smoke and fire, from a variety of DreamWorks Animation feature films. He will discuss in general terms the kinds of techniques used to achieve particular visual effects. Finally, Henderson will show a detailed breakdown of the dam-breaking scene from Madagascar: Escape 2 Africa, demonstrating how different elements of key frame animation, simulation, and rendering are combined in a real production shot.

  • December 1, 2009
    A Quick and Easy Way to Parallelize a Legacy Codebase with Intel® Threading Building Blocks (TBBs)
    Speaker: Bernard Laberge, Avid, Senior Principal Engineer (Bio)

    Bernard Laberge is a senior principal engineer in the video editors division at Avid. During his seven years with the company he has been actively involved in the replacement of the legacy video processing engines used by Avid editors with a common hardware-abstracted, component-based video processing engine currently running on the CPU with SIMD optimized code, GPU, and dedicated hardware.

    Abstract:
    Learn how to overcome the limitations of a thread-based scheduler, including dealing with the absence of recursive parallelism support and the inefficient handling of unbalanced processing load. Bernard Laberge addresses how Avid resolved the expensive refactoring of their thread-based scheduler into a task-based solution by choosing Intel® Threading Building Blocks (TBBs). He explores how Avid was able to easily integrate the Intel TBBs into their video editor applications and more than 5 million lines of code.

  • December 15, 2009
    How to Use Intel® Parallel Studio to Streamline Code Development in a Multicore Environment
    Speaker: Matt Dunbar, Director for Performance Technology, SIMULIA (Bio)

    Matt Dunbar is the director for performance technology at SIMULIA. Since joining the company in 1993, he has worked on parallelization of the Abaqus suite of products, initially for shared memory architectures and more recently for distributed memory architectures. Dunbar has also been intimately involved in selecting both the hardware and software tools used in the development of the Abaqus product line.

    Abstract:
    Resolve elusive, costly multithreading errors quickly and efficiently with Intel® Parallel Studio. While many coding problems that lead to bugs in software applications are typically straightforward logic errors, errors in managing memory and in multithreading code can sometimes take weeks to months to diagnose and fix. Matt Dunbar explores how and why taking advantage of multicore processors through multithreaded code is critical for compute-intensive applications. While spotlighting his work on SIMULIA's Abaqus finite element solver, Dunbar addresses the need for multicore execution and shares his experiences using Intel Parallel Studio to streamline code development in a multicore environment.