Report Navigation
Many scientific applications are very floating-point calculation intensive. In comparison with traditional commodity CPUs, high-end graphics processing units (GPUs), dedicate a much higher portion (approximately 35%) of their logic to floating-point operations that also are essential for fast graphics operations. GEaRS staff worked closely with PeakStream, a supplier of software APIs to facilitate numerical computations on such graphics, and Dr. John Liechty, associate professor of marketing, to employ high-end GPUs in numerical analyses of market data. Work was promising until Google purchased PeakStream and withdrew the technology from the consumer market.
The Cell Processor is an exciting new processor from IBM designed to handle large amounts of floating-point computation by utilizing an 9-core CPU design. One of the cores acts as the main core and the other 8 cores act as offload engines to the main core. It is estimated that the Cell Processor will reach around 256 gigaFLOPS (or GFLOPS - one billion floating-point operations per second). GEaRS staff worked with Dr. Padma Raghavan, professor of computer science and engineering, and director of Penn State's Institute for Computational Science, and IBM to develop a successful Shared University Research (SUR) award to bring Cell technology to Penn State for research into tools for improving the scientific programming environment for the Cell architecture. IBM's SUR program awards unique IBM technologies to university researchers to facilitate investigations and development of mutual benefit.