The Graph-Massivizer project consortium is happy to announce the official start of this European initiative, funded by the European Commission under the Horizon Europe research and innovation programme. Graph-Massivizer aims at delivering open-source and commercial solutions that drive green digital transformation across use cases in finance, manufacturing, environment protection and exascale computing.

Leveraging graph data through an efficient and scalable digital infrastructure driving green digital transformation

Over the last decade, graphs have made great advances in making data findable, accessible, interoperable, and reusable. For many organisations, they have become a key instrument for extracting meaningful insights that support timely, high-impact decisions. On a larger scale, graphs are becoming crucial to innovation, competition, and prosperity. They help derive trustworthy insights to create sustainable communities and support digital transformation with better, more profitable, greener products and services. However, current graph processing platforms come with various limitations, ranging from high energy consumption and inefficiency and lack of support for diverse workloads, models, languages, and algebraic frameworks to the difficulty of use for non-experts.

Graph-Massivizer addresses these challenges by delivering an integrated toolkit to support a climate-neutral and sustainable economy based on graph data. The project partners will develop five open-source software tools for high-performance, scalable, and sustainable graph processing, as well as an enterprise-class commercial version based on the metaphactory knowledge graph platform that tightly integrates the tools in an easy-to-use-and-deploy offering to reach a broader market share.

Cineca will take part in the project with its expertise in developing Digital Twins and managing big amounts of data. 

Data Center Digital Twin is a solution developed by Cineca and the University of Bologna that provides a virtual representation of the world’s fourth-fastest supercomputer Leonardo. Leonardo’s digital massive graph representation describes complex spatial, semantic, and temporal relationships between the monitoring metrics, hardware nodes, cooling equipment, and software, which are difficult to capture and express otherwise.

“Once operative, Leonardo will generate over 10 million metrics and petabytes of data that require AI analytics on massive graphs to extract operational insights for improved science throughput. The information in such a large volume of data is essential for understanding and optimising the efficiency and sustainability of future modern supercomputers operating at exascale performance,” says Andrea Bartolini, assistant professor at the University of Bologna.

More information in the press release.