CRM Researchers Are Developing a New Efficient Data Compression Method

Researchers from the Centre de Recerca Matemàtica (CRM) are working on an innovative project called DYSEDAS, which aims to develop an efficient method for storing large amounts of data using techniques from symbolic dynamical systems. The project’s goal is to minimize data sizes without any loss of information. Initially, the focus is on compressing image files, which typically require substantial storage and bandwidth. Leading the project are Lluís Alsedà, a professor at the Universitat Autònoma de Barcelona and Director of the CRM, and David Romero, Director of the CRM Knowledge Transfer Unit.

David Romero explains that the main idea “is to develop a compression system that matches the level of existing compression systems, without requiring high levels of computation, making it more accessible for everyone to use.”

The tech industry’s growth has led to an increased demand for data storage, resulting in significant environmental consequences. Data centres, which store and process vast amounts of information, consume large amounts of electricity and contribute to CO2 emissions. By reducing data sizes, DYSEDAS aims to lower the energy requirements for data storage and transmission, thus reducing greenhouse gas emissions and environmental impact.

“Maintaining data is expensive because large data centers consume a lot of energy” adds Romero. “The goal of DYSEDAS is not just to create a compression algorithm specifically for images but to develop one that can handle any file type. There are algorithms that work better for images than text, and our aim is to address this variability.”

Moreover, the project offers potential economic benefits by lowering the need for extensive data storage infrastructure. This is especially beneficial for sectors like academia, healthcare, and public services that rely heavily on large volumes of data. For instance, the healthcare industry generates vast amounts of imaging data that require efficient storage solutions. DYSEDAS’s technology can make these processes more sustainable and cost-effective.

A Broader Vision for Sustainability

Projects such as DYSEDAS hope to contribute towards a more sustainable future where technology innovation does not come at the expense of our natural resources and ecosystems. By minimizing the environmental impact of managing the large volumes of data that nowadays flood storage facilities, the project supports global efforts to reduce carbon footprints. Additionally, the method’s universal applicability means it can be extended to other types of data beyond images, potentially transforming data storage across various domains.

The project also aligns with the goals of the FAIR/Open Science movement, promoting the use of scientific and technical data for innovation and modernization. By enhancing data management and distribution, DYSEDAS supports public sector agencies involved in critical areas such as weather, climate, and emergency services.

Romero highlights the importance of reproducibility in science, stating, “For science to be fair, experiments must be reproducible. The results obtained from the data should be reproducible, meaning that existing data should be usable in an open manner. If there is a lot of data, to conduct new experiments, this data must be stored compactly, accessible, and without any data loss using the compression algorithm.”

DYSEDAS is funded by the Spanish Ministerio de Ciencia, Innovación y Universidades as part of the Transición ecológica y transición digital call. This funding initiative aims to promote R&D activities to increase the competitiveness and international leadership of Spanish science and technology through the generation of scientific knowledge, quality research, and the development of technologies.

Digital transition projects such as DYSEDAS aim to enhance the infrastructures, skills, and technologies needed for a digital economy and society, fully leveraging the synergies and opportunities of new technological developments and data management, while prioritizing people and their digital rights.