Remote Sensing Data Analysis
using HTC/HPC Systems
Dr. Francesco Nattino
Dr. Francesco Nattino has a master's degree in Chemistry from the University of Milan. His interest in the use of computer simulations to answer research questions brought him to specialize in Theoretical and Computational Chemistry, a field in which he carried out his PhD at Leiden University. After working as a postdoctoral researcher in the Materials Science department at EPFL, Switzerland, Francesco joined the Netherlands eScience Center as a Research Software Engineer in 2020. In such a position, he has combined his experience in HPC/HTC with a newly developed passion for remote-sensing and geospatial data applications. Beside his involvement in research projects, he is also active in teaching: he is a Carpentries instructor and one of the developers of the Carpentries Incubator lesson “Introduction to Geospatial Raster and Vector Data Using Python”.
Dr. Meiert Willem Grootes
Dr. Meiert Willem Grootes is a Senior Research Software Engineer at the Netherlands eScience Center. After obtaining a PhD in astrophysics from the University of Heidelberg he pursued research on galaxy evolution as an independent postdoctoral fellow at the Max-Planck Institute for Nuclear Physics and ESA, including work on data analysis pipelines for satellite images, machine learning based galaxy classification techniques, radiative transfer models, and statistical analysis. At the eScience center he is involved in projects from the Environment & Sustainability and Natural Sciences & Engineering sections, with a focus on earth observation, data storage and access, and machine learning. He is further interested in HPC/HTC and hardware acceleration, and their application in research.
Dr. Pranav Chandramouli
Dr. Pranav Chandramouli is a Research SoftwareEngineer at the Netherlands eScience Center. He has a PhD in the field ofapplied mathematics from INRIA, France, where he developed the 4DVar suite ofmodels for variational data assimilation in turbulent flows. He has abackground in CFD, climate models, and turbulence. At the eScience center, he worksin the Environment and Sustainability section on projects related to earthobservation, machine learning, and statistical analysis amongst othertopics. He is also interested in the fields of quantum computing, artificialintelligence, GPUs and HPC/HTC.
Remote Sensing Deployable Analysis Environment
Remote sensing (RS) data in general, and Earth observation (EO) data, have become a mainstay in fields ranging from the geosciences to ‘green’ life sciences, agriculture, and even social sciences, as well as an invaluable tool in defining policy. In response, a community driven software ecosystem has evolved to support exploitation of these data.
With the volume of these data increasing incessantly, existing and future workflows often must be scaled up beyond the computational and storage resources available in workstations. In this regard, solutions using high-throughput and high-performance computing (HTC/HPC) systems, as an additional alternative to cloud solutions, are of relevance for the academic community. Offering full control over available hardware, software, and data, these systems are excellently suited to highly customised academic workflows and can readily support the migration of existing workflows. Furthermore, they are generally available through national infrastructure providers on a merit-driven no-cost basis.
This session will cover the basic tenets of the use of large academic computing resources, and introduce participants to a Dask-based ecosystem, familiarising them with the use of the Remote Sensing Deployable Analysis environment framework to scale EO and RS data analysis using HTC/HPC systems and associated storage resources. The session will cover the tools for data access, retrieval and storage, and demonstrate the scaling up of processing and analysis workflows focused on EO datasets. Participants will perform hands-on research using the RSDAT framework on an HTC/HPC system.
- Basic understanding of Unix command line and shell, the python programming language, and the geospatial python ecosystem*
- Affinity to high-performance computing
- Interest in scaling EO workflows
* We strongly advise participants to familiarize themselves with the contents of the (nascent) carpentries geospatial python lesson
Dr. May Casterline
Dr. May Casterline is a data scientist/image scientist/software developer with a background in satellite and airborne imaging systems. Her research interests include deep learning, hyperspectral and multispectral imaging, innovative applications of machine learning approaches to remote sensing data, multimodal data fusion, data workflow design, high performance computing applications, and creative software solutions to challenging geospatial problems. She holds a PhD and Bachelors of Science in Imaging Science from Rochester Institute of Technology, with a focus on remote sensing. In industry she has acted as a product owner, technical lead, lead developer, and image scientist on both research initiatives and development projects. As a Senior Solutions Architect at NVIDIA, Dr. Casterline works with both industry and government to help enable developers, engineers, data scientists and analysts integrate artificial intelligence and GPU-accelerated solutions into their workflows and products.
Alison Lowndes (Senior Scientist, Global AI | NVIDIA). Joining in 2015, Alison spent her first few years as a Deep Learning Solutions Architect and is now responsible for applied Artificial Intelligence both around the globe and off Earth, in Space. A mature graduate in AI, Alison combines technical and theoretical computer science with astrophysics & over 25 years of experience in international project management, entrepreneurial activities and the internet. She consults on a wide range of applications, including planetary defence with NASA, ESA & the United Nations and works closely with world governments advising them on how to harness AI for economic growth, national security & climate action, using NVIDIA’s GPU Computing platform.