Laura Darriba-Pol | Searcharxiv

arXiv2026

Bringing computation to the data: A MOEA-driven approach for optimising data processing in the context of the SKA and SRCNet

The Square Kilometre Array (SKA) will generate unprecedented data volumes, making efficient data processing a critical challenge. Within this context, the SKA Regional Centres Network (SRCNet) must operate in a near-exascale environment where traditional data-centric computing models based on moving large datasets to centralised resources are no longer viable due to network and storage bottlenecks. To address this limitation, this work proposes a shift towards distributed and in-situ computing, where computation is moved closer to the data. We explore the integration of Function-as-a-Service (FaaS) with an intelligent decision-making entity based on Evolutionary Algorithms (EAs) to optimise data-intensive workflows within SRCNet. FaaS enables lightweight and modular function execution near data sources while abstracting infrastructure management. The proposed decision-making entity employs Multi-Objective Evolutionary Algorithms (MOEAs) to explore near-optimal execution plans considering execution time and energy consumption, together with constraints related to data location and transfer costs. This work establishes a baseline framework for efficient and cost-aware computation-to-data strategies within the SRCNet architecture.

cs.DC↗

arXiv2025

Classification of HI Galaxy Profiles Using Unsupervised Learning and Convolutional Neural Networks: A Comparative Analysis and Methodological Cases of Studies

Hydrogen, the most abundant element in the universe, is crucial for understanding galaxy formation and evolution. The 21 cm neutral atomic hydrogen - HI spectral line maps the gas kinematics within galaxies, providing key insights into interactions, galactic structure, and star formation processes. With new radio instruments, the volume and complexity of data is increasing. To analyze and classify integrated HI spectral profiles in a efficient way, this work presents a framework that integrates Machine Learning techniques, combining unsupervised methods and CNNs. To this end, we apply our framework to a selected subsample of 318 spectral HI profiles of the CIG and 30.780 profiles from the Arecibo Legacy Fast ALFA Survey catalogue. Data pre-processing involved the Busyfit package and iterative fitting with polynomial, Gaussian, and double-Lorentzian models. Clustering methods, including K-means, spectral clustering, DBSCAN, and agglomerative clustering, were used for feature extraction and to bootstrap classification we applied K-NN, SVM, and Random Forest classifiers, optimizing accuracy with CNN. Additionally, we introduced a 2D model of the profiles to enhance classification by adding dimensionality to the data. Three 2D models were generated based on transformations and normalised versions to quantify the level of asymmetry. These methods were tested in a previous analytical classification study conducted by the Analysis of the Interstellar Medium in Isolated Galaxies group. This approach enhances classification accuracy and aims to establish a methodology that could be applied to data analysis in future surveys conducted with the Square Kilometre Array (SKA), currently under construction. All materials, code, and models have been made publicly available in an open-access repository, adhering to FAIR principles.

astro-ph.GA↗