ESAPCA: Distributed Computing for Large-Scale Data Analysis

The ESAPCA project developed parallel, GPU-accelerated implementations of SVD and related data science techniques that could work on high-performance computing systems. The solution integrates these analysis techniques into the Heat research software library, allowing scientists to process massive datasets across multiple computing nodes whilst maintaining compatibility with familiar Python tools like NumPy and SciPy. This approach enables researchers to analyse extremely large datasets that were previously impossible to process on conventional computing systems.