Fresh Outlook on Numerical Methods for Geodynamics. Part 1: Introduction and Modeling

Since 2015 much has developed in geodynamical modeling because of the arrival of Big Data. We present in two parts an overview of numerical techniques but also a scan of the new opportunities in this age of Big Data and prepare the community for the coming decade, the roaring twenties, when Data Analytics will reign. We begin with a review of traditional numerical methods (Part I), followed by a survey of the current techniques used for data analytics and high-performance computing (HPC) (Part II). Our aim is to cover topics of machine learning, neural networks and deep learning, unsupervised learning as well as the role that HPC will play in the Big Data era, especially in hardware of various calibers. Finally, we will address the need for education of students and professionals, in particular, on the use of the emerging programming languages and the importance of scientific software communities.