 Computational Sciences Campaign is a cross-cutting campaign. We work with the subject matter experts in the four fundamental campaigns to be able to find solutions using a high-performance computing and other computing methods. We have multiple focus areas within the Computational Sciences Campaign, the first of which is predictive sciences. Predictive science is really all about how do you enable quick decisions in a complex environment. So for the soldier, how can we leverage both predictive models as well as sensor information that may come in, historical data, combine that together to make a decision? If you look to the future of computing, what you'll see is that we're quickly approaching a wall. It's a physical wall as to how small we can make transistors and how much power we can fit into an area of a computer chip without it melting. We still want to develop systems that go fast. The question is how do we do that? And the answer is you need to progress for individual types of problems. When we say or talk about emerging architectures, we really mean specialized computers that work in a very unique kind of way to solve a specific problem very efficiently. We're interested in bringing advanced computing capabilities to the soldier in the battlefield using devices that operate within extreme size, weight and power constrained environments. Machine learning and artificial intelligence. So for example, your phone may last a day whereas the architectures we're looking at may last up to a month. The last area is new methods for data intensive sciences where we develop methods to manage process and also to reduce the large volumes of data that are coming from the battlefield and also from experiments within the laboratory to be able to understand those. Some of the building blocks, whatever we do have are rightly aligned to really support some of the ERAs. Some of the ERAs we're going to make a big impact are in the areas of artificial intelligence and machine learning. The computing architectures and the edge with low power is going to create a scenario which is essential for the AI machine learning ERA. Similarly, with the cyber and EW plot scale modeling is going to be an essential area for research component under the ERA. We aim to ensure that the Army maintains information dominance. The fastest device is the best algorithms and is able to deliver the appropriate information to the decision maker the quickest.