Towards Realistic Performance Bounds for Implicit CFD Codes

The performance of scientific computing applications often achieves a small fraction of peak performance [7,17]. In this paper, we discuss two causes of performance problems— insufficient memory bandwidth and a suboptimal instruction mix—in the context of a complete, parallel, unstructured mesh implicit CFD code. These results show that the performance of our code and of similar implicit codes is limited by the memory bandwidth of RISC-based processor nodes to as little as 10% of peak performance for some critical computational kernels. Limits on the number of basic operations that can be performed in a single clock cycle also limit the performance of “cache-friendly” parts of the code.