When GIS Hits the Wall: Scaling Flood Modeling with High-Performance Computing (HPC)
2026-08-30 , 101

Modern GIS workflows increasingly exceed desktop limits. This workshop shows when and why HPC becomes essential for GIS, demonstrating how open-source tools can scale flood and environmental modeling using parallel and distributed computing.


Geospatial workflows are rapidly growing in scale and complexity, driven by high-resolution DEMs, long satellite time series, climate datasets, and computationally intensive models such as flood and inundation simulations. While open-source GIS tools have matured significantly, many real-world problems now exceed the limits of desktop-based processing.

This workshop introduces High-Performance Computing (HPC) as a natural extension of modern GIS, not as a replacement. Using flood and environmental modeling as core examples, the session demonstrates where traditional GIS workflows struggle and how parallel and distributed computing can unlock new possibilities using fully open-source tools.

Participants will learn HPC concepts through familiar GIS analogies and see how tools like QGIS, GDAL, Python, Dask, and ANUGA Hydro can be integrated with HPC systems. Through live demonstrations and guided workflows, the workshop shows how large raster processing, time-stepped simulations, and spatio-temporal analytics can scale efficiently on HPC infrastructure.

The workshop emphasizes practical adoption, showing how GIS users can continue working in their existing open-source ecosystem while leveraging HPC for heavy computation. By the end of the session, participants will understand when HPC is necessary, how it fits into GIS pipelines, and how to begin using it for real-world geospatial challenges.


Level of the workshop: 2 - intermediate Pre-requirements for attendees:

Basic understanding of GIS concepts (rasters, vectors, projections)
Familiarity with QGIS or Python-based GIS tools is helpful but not mandatory

What skills do participants require to have?:

Basic understanding of GIS concepts (rasters, vectors, projections)
Familiarity with QGIS or any open-source GIS tool
Basic command-line usage (helpful but not mandatory)
Introductory Python knowledge is beneficial but not required
No prior experience with HPC or parallel computing is required

I am a Scientist at C-DAC Pune, under the Ministry of Electronics and Information Technology (MeitY), Government of India. My work focuses on large-scale geospatial analytics, flood and environmental modeling, and integrating high-performance computing (HPC) with open-source GIS frameworks. I actively work on scalable simulation workflows, geospatial AI, and bridging desktop GIS with supercomputing for real-world decision support.