Q+A CENTER

System integration

The GPUs are intended as the graphical processor companion in every PC. Therefore, most modern PCs can accommodate the presence of a modern GPU, with respect to power supply and cooling. Consequently, all reconstruction hardware that fits in the same {cooling, power supply} envelope can be hosted in the same host PC. The Cell Accelerator Board (CAB)—a high performance accelerator card based on the Cell BE processor--has been designed to fit into this envelope and can be hosted in any modern PC. FPGA-based boards are subject to the concept of the designer. FPGAs traditionally draw less power than high-clocked devices, such as a GPU or the CBE processor, and are easier to cool.

Conclusion

The implementation of a cone-beam reconstruction algorithm was investigated for various platforms. All basic building blocks, i.e. GPU, FPGA and cell, are available and can deliver the appropriate image quality, however with varying degrees of effort. Depending on the evaluation criteria, the optimal choice between FPGAs, GPUs, multi-core based PCs and the cell processor may differ. However, the Cell Broadband Engine offers a fully programmable architecture, accessible from high-level programming languages such as C. There also are complete software development suites that significantly simplify the programming steps necessary to implement a given application on a multicore architecture like the cell and help programmers keep a high level of productivity. The high-processing density that the cell processor offers makes it possible to solve most of the reconstruction problems within the same computer. A comparable processing power can be achieved with traditional blade cluster systems, at the expense of a higher application complexity and higher system failure rates.