With the amount of high resolution earth observation data available it is not feasible anymore to do all analysis on local computers or even local cluster systems. To achieve high performance for out-of-memory datasets we develop the YAXArrays.jl package in the Julia programming language. YAXArrays.jl provides both an abstraction over chunked n-dimensional arrays with labelled axes and efficient multi-threaded and multi-process computation on these arrays.
In this contribution we would like to present the lessons we learned from scaling an analysis of high resolution Sentinel-1 time series
data. By bringing a Sentinel-1 change detection use case which has been performed on a small local area of interest to a whole region we test the ease and performance of distributed computing on the European Open Science Cloud (EOSC) in Julia.