Search results

This aim of this work is to investigate different modelling approaches for air-cooled data centres. The study employs three computational methods, which are based on finite element, finite volume and lattice Boltzmann methods and which are respectively implemented via commercial Multiphysics software, open-source computational fluid dynamics code and graphical processing unit-based code developed by the authors. The results focus on comparison of the three methods, all of which include models for turbulence, when applied to two rows of datacom racks with cool air supplied via an underfloor plenum.

This paper studies thermal airflows in a data centre by applying different numerical simulation techniques that are able to analyse the thermal airflow distribution for a simplified layout of datacom racks in the presence of a computer room air conditioner.

Good quantitative agreement between the three methods is seen in terms of the inlet temperatures to the datacom equipment. The computational methods are contrasted in terms of application to thermal management of data centres.

The work demonstrates how the different simulation techniques applied to thermal management of airflow in a data centre can provide valuable design and operational understanding. Basing the analysis on three very different computational approaches is new and would offer an informed understanding of their potential for a class of problems.

By applying targeted graph algorithms, the method used by the authors enables effective prediction of user interactions and thus fulfils the complex requirements of modern recommender systems. This study sets a new benchmark for multidimensional recommendation strategies and offers a path towards more advanced and user-centric models.

To improve multidimensional data recommendation systems, multiplex graph structures are useful to capture various types of user interactions. This paper presents a novel framework that uses a graph database to compute and manipulate multiplex graphs. The approach enables flexible dimension management and increases expressive power through a specialised algebra designed for multiplex graph manipulation.

The authors compare the multiplex graph approach with traditional matrix methods, in particular random walk with restart, and show that the method not only provides deeper insights into user preferences by integrating scores from different layers of the multiplex graph, but also outperforming matrix-based approaches in most configurations. The results highlight the potential of multiplex graphs for developing sophisticated and customised recommender systems that significantly improve both performance and explainability.

The study provides a formal specification of a multiplex graph construction based on interaction and content-based information; and the study also developed an algebra dedicated to multiplex graphs, enabling robust and precise graph manipulations necessary for effective recommendation queries. The authors implement these algebraic operations within the Neo4j graph database system with a thorough analysis and experimentation with three different data sets, benchmarked against traditional matrix-based methods.