Monitoring geospatial workflows with precision for better performance, resource utilization, and reproducibility

Description

The rapid growth of geospatial data from satellites, drones, sensors, and social platforms has made efficient analysis increasingly challenging. Modern workflows often involve complex software and hardware stacks, where computational inefficiencies can lead to wasted resources, high costs, and limited reproducibility. This presentation highlights the importance of monitoring and benchmarking geospatial workflows to optimize performance and resource utilization. It introduces best practices for experimental design, workflow management, data handling, performance metrics, and transparent reporting to ensure reproducibility and reliability. A key focus is on Geobench, a new open-source framework designed to monitor and benchmark geospatial tasks systematically. Geobench supports multiple execution environments, collects detailed system and workflow metrics, and generates reproducible, human- and machine-readable reports. By adopting structured monitoring and benchmarking, researchers and practitioners can make informed decisions about computing strategies, improve methodological rigor, and contribute to open and scalable geospatial science.

Period	4 Sept 2025
Event title	EO Summer School 2025: Data Science for Earth Observation: Hands-on Training in Python, R and Julia
Event type	Workshop
Location	Wageningen, NetherlandsShow on map
Degree of Recognition	International