high performance computing on graphics processing units: hgpu.org

The value of Prognostics and Health Management (“PHM”) is best understood by looking at what happens when PHM is not utilized. In the growing field of PHM, the mistakes of others often provide us with valuable real world lessons that should fuel greater development, understanding and use of PHM. The science of PHM should play an ever increasing role in the management of commercial risk associated with new product development. Taking liberty with George Santayana’s famous quote: “Those who cannot remember the past are condemned to repeat it,” a student of PHM might say, “Those who ignore PHM are condemned to fix the failure rather than prevent it. ” At least S300 million is the cost of trying to fix the many problems caused by Nvidia’s defective graphics processing units (“GPU”). As consumers began observing and reporting failures in their computers to companies like Hewlett-Packard, Asus, Loshiba, Dell, and others, the absence of effective PHM programs caused delay in understanding the problems, the failure mechanisms, the root cause and the available corrective actions. Lhese delays resulted in continued production and sale of defective units, ineffective solutions, consumer and securities lawsuits, and dramatically increased costs. As the American saying goes, “An ounce of prevention is worth a pound of cure.”