> The question is, what are those compute bound workloads? I'm not persuaded that there are too many of them anymore, and the real bottleneck for some time with

> The question is, what are those compute bound workloads? I'm not persuaded that there are too many of them anymore, and the real bottleneck for some time with most problems has been I/O. This even extends to GPUs where fast memory makes a huge difference. The bottlenecks in the problems themselves shouldn't be underestimated though. Some types of problems are intrinsically difficult or outright impossible to reformulate so as to take advantage of vectorized processing. That being said, there are a lot of other problems which do. I'm quite enthusiastic about this.