Reflections on science, technology, and computing — leavened by personal experience


Our run of good luck is over. We need new, integrated design methodologies and rich collaborations that consider end-to-end design and capability. In many ways, this is a return to the past.

We can continue down our hand-crafted application development path for petascale systems, and it will work; the evidence to date is clear. Given enough investment of time and money, we will be successful, but I am not convinced we are on the most effective path. (Aerodynamics is not mandatory on cars either, but it surely…