The Purpose of Computing is Insight, not Numbers, Mark Miller


HPC has the potential to revolutionize the scientific method by providing a third pillar, simulation, which helps to inform the other two, theory and experimentation. But developing effective and truly predictive simulation science capabilities is a complex, multi-disciplinary and highly collaborative team effort incorporating expertise from numerous disciplines including domain sciences (e.g. cosmology, fusion, medicine, engineering, etc.), mathematics, numerical methods, computer science, software quality engineering, devops, visualization, technical writers and even artists and psychologists. In our path through the 2022 ECP Annual Meeting, we will be keeping an eye out for the range of disciplines we see coming together in any given presentation and understanding what our relationship to those disciplines is, both currently as well as what gaps we might need to fill to improve those relationships that interest us. Likewise, we will be looking for metaphors to our own experiences and lives that help us understand and relate to the activities of others.

Short Biography:

Mark C Miller has participated in the development of a number of scientific database and data modeling technologies including Silo, ASCI-DMF, HDF5, ITAPS and most recently the MACSio scalable I/O proxy application. Mark has been the lead developer of Silo since the late 90’s supporting scalable I/O requirements of LLNL HPC simulation codes including ALE3D, Kull, Ares, LASNEX, Overlink and VisIt. Mark’s expertise includes data models and their impact on software interoperability, high performance I/O, Software Quality Engineering (Smart Libraries) for HPC libraries and, more recently, in-situ data exchanges and transitioning from a push- to a pull-model for I/O. In his view, a majority of his activities and experiences over the last 20 years of HPC are best described as those of a “Virtual Ramrod”.


I would like to help SRP-HPC participants find their passion(s) in HPC and see through the often overwhelming details to where they might find a place and interest in an HPC career.