CAD/CAM Corner

Auto Industry Bailout: Pro, Con And Undecided

Unlike the recent presidential election where I found it hard to fathom that there could be actual “undecideds,” I admit I fall into that embarrassingly squishy category when it comes to the possibility of a bailout (or a bridge loan, as the auto makers like to call it) for the big three car companies.

On one hand, having covered manufacturing, and to some extent, the automotive supply chain, for years, I know first hand the damage that will be inflicted beyond Detroit if we let this problem go unchecked and the industry go under. Beyond the shock and confidence crushed resulting from a bankrupt U.S. auto industry, the very intricate and ample supply chain of small and mid-sized companies, some of which have been around for decades, will start to crumble. No doubt, millions of jobs will be lost, manufacturing infrastructure will sit idle and many of these companies will inevitably go under. Here, is the case, my one side argues, for providing some sort of government assistance. Call me crazy, but I don’t see the difference here between this situation and that of the Wall Street bailout. The government lent a hand to the rich, white collar industry, why not do the same for bread and butter manufacturing?

On the other hand, this crisis is not exactly a new development. GM chief Rick Wagoner and his CEO peers told the Senate Banking Committee that the automotive industry was in dire straits mostly due to the recent and precipitous economic downturn. They claimed the industry was on the brink of a transformation, and the recent tight lending conditions and free fall in consumer spending is what’s forcing the industry over the edge.

Wait a second. Hasn’t the automotive industry has been struggling publicly for years, and maybe privately, for decades, trying to move beyond its infatuation with gas-guzzling SUVs and pickup trucks to pursue innovative alternative energy technologies? Haven’t U.S. car companies been desperately trying to reduce their product development cycles, get a handle on lean manufacturing and generally, be more competitive with foreign cars makers? The truth is, this isn’t some bomb dropping out of the blue—this is a firestorm that’s been underway for some time. It’s just that the auto industry hasn’t been able to accomplish the massive directional changes fast enough with any kind of sustainable success.

So I take exception to the auto execs tying the immediate need for government assistance to the recent economic crisis. It’s the sticking point that makes me think maybe we should take a pass on any bailout measure. This industry made it’s own bed. Perhaps, we have to hang tough and let them lie in it.

Now you see why I’m so squishy?

TOP500 Supercomputer List Debuts

I bet you’re dying to know the know status of the world’s TOP500 supercomputers. Well, the 32nd edition has just been issued and there’s a bit of an upset: IBM’s 1.105 petaflop/s Roadrunner supercomputer, installed at the Los Alamos National Laboratory, has managed to maintain its top spot, narrowly fending off a challenge by a Cray XT5 supercomputer at Oak Ridge National Laboratory, affectionately nicknamed Jaguar, which was recently upgraded this year and expected to nudge out its competitor. That system, only the second to break the petaflop/s barrier, posted a top performance of 1.059 petaflop/s in running the Linpack benchmark application. One petaflop/s represents one quadrillion floating point operations per second.

Both of the systems, along with five other of the top supercomputers, reside at national laboratories operated by the U.S. Department of Energy (DOE).  Nine of the top 10 supercomputers are located in the United States, and the most powerful system outside of U.S. borders is the Chinese-built Dawning 5000A at the Shanghai Supercomputer Center. It is the largest system operating with Microsoft Windows HPC 2008.

Some other highlights from the November 2008 list:
* A total of 379 systems (75.8 percent) are now using Intel processors, virtually unchanged from six months ago (375 systems, 75 percent).

* Quad-core processor based systems have taken over the TOP500, deployed in 336 systems. Dual-core processors are used in 153 systems, and only four systems still use single core processors. IBM’s advanced Sony PlayStation 3 processor with nine cores is gaining steam—seven systems on the list have tapped that platform.

* HP took over the lead in systems with 209 systems (41.8 percent) over IBM with 188 systems (37.6 percent).  IBM had 210 systems (42.0 percent) six months ago, compared to HP with 183 systems (36.6 percent).

* The United States remains the leading consumer of HPC systems, with 291 of the 500 systems (up from 257). Comparatively, Europe claims 151 systems, down from 184 and Asia remains unchanged at 47 systems.

NVIDIA Breaks The GPU Bank with Quadro FX 5800

Looking to break the bank with graphics processing horsepower? NVIDIA’s  just-released Quadro FX 5800, the flagship model in its professional GPU series, and the Quadro Plex 2200 S4 Visual Computing System, a large-scale server system based on that platform, aims to up the ante, both in graphics visualization and performance.

The building block of the new entrants is the Quadro FX 5800, a $3,499 graphics card that with its 4GB of graphics memory and up to 240 CUDA programmable parallel cores (the equivalent of 1 teraflops of processing power) delivers the performance and scalability to visualize and interpret massive datasets that were previously unattainable on workstation-class graphics boards, according to Jeff Brown, NVIDIA’s general manager of professional solutions. Pitched for oil and gas exploration, medical imaging and high-end styling and industrial design applications, the Quadro FX 5800 can bring together and render an entire assembly interactively, with higher degrees of precision and realism previously unavailable, Brown said.

For companies looking to blend computational and graphics work simultaneously—for running crash analysis on a model, for example, making a change and then running the simulation again to see how the change impacts the basic integrity of a model–NVIDIA’s Quadro Plex 2200 comes into play. The system, available in a standard 1U form factor for large-scale server deployment, taps the power of four Quadro FX 5800 GPUs to tackle the most challenging visual supercomputing problems. The Quadro Plex 2200 will be available in December for $14,995.