Main Content Region

Ongoing Research Projects supported by Research IT

Listing of project codes and abstracts, describing work undertaken which use the resources of the compute clusters hosted by the Research IT team.

New codes can be obtained by filling in this resource application form.

Search Results
Showing 10 of 419 Results
Project Title Stastistical optimisation of Functional Connectivity Metrics
Project Code HPC_18_01030
Principal Investigator Assistant Professor Clare Kelly
Start Date 2018-07-01
End Date 2018-09-30
Abstract The aim of this project is to use various resampling techniques to improve the reliability and accuracy of metrics of functional connectivity as established by fMRI. MRI and resting-state fMRI (rfMRI)Data is taken from open source CORR datasets. By applying bootstrap resampling techniques, this project looks to shorten the time necessary to scan a patient and establish functional connectivity metrics. It also looks to improve the accuracy and reliability of these established metric. We use data from participants with multiple scans to see if resampling has a concrete effect on our ability to identify subjects using functional connectome fingerprinting.
Project Title Cellular Signal Mobility
Project Code HPC_18_01029
Principal Investigator Assistant Professor Martina Kirchberger
Start Date 2018-07-18
End Date 2019-07-20
Abstract Using large amounts of cellular user data, we are determining measures of mobility and frequency of mobility in African and Asian countries. Geospatial models in R, primarily, will be used on user cellular microdata to determine their frequency of movements, locations, and other spatial modeling.
Project Title Many-body theory of antimatter interactions with atoms, molecules and condensed matter
Project Code HPC_18_01028
Principal Investigator Dr Charles Patterson
Start Date 2018-06-27
End Date 2021-02-01
Abstract This research programme will develop state-of-the-art many-body theory and its computational implementation to enable the most accurate calculations of positron and positronium interactions with atoms, molecules and condensed matter. Many-body theory is a powerful method that provides a natural, intuitive and systematic account of important positron-molecule and electron-positron correlations. This work will aim to elucidate the role of the strong correlations in positron-molecule and positron-condensed matter interactions in particular, providing fundamental insights that will enable the most accurate interpretation of fundamental atomic physics experiments, positron-based materials science techniques, and the development of next-generation antimatter-based technologies (e.g., positron traps and positron emission tomography).
Project Title DFT calculations over copper oxide/cerium oxide catalysts in preferential CO oxidation (CO-PROX)
Project Code HPC_18_01027
Principal Investigator Assistant Professor Max Garcia Melchor
Start Date 2018-06-20
End Date 2019-06-20
Abstract Preferential CO oxidation (CO-PROX) is a promising strategy to allow direct introduction of hydrogen produced by steam reforming into fuel cells for energy applications. Mixed copper and cerium oxide catalysts are materials that display an excellent synergistic performance towards CO-PROX in H2-rich gas mixtures, where the suppression of H2 oxidation is a key requirement. Despite extensive experiments and thorough characterisation studies reported over decades, a detailed computational investigation of this complex reaction environment is still lacking. Hence, this project aims to use advanced density functional theory (DFT) methods to reach a fundamental understanding of the complex CO-PROX reaction. The results derived from this project are expected to provide a unique insight into: (i) the range of CO oxidation selectivities observed by with different copper oxidation states; (ii) the abundance and stability of surface reaction intermediates based on experimental kinetic studies; (iii) the oxygen diffusion mechanism and rates over the cerium oxide support.
Project Title Excitonic Density-Functional Theory
Project Code HPC_18_01025
Principal Investigator Prof David O'Regan
Start Date 2018-06-04
End Date 2018-09-10
Abstract State-of-the-art methods for calculating neutral excitation energies are typically demanding and limited to single electron-hole pairs and their composite plasmons. I will introduce excitonic density-functional theory (XDFT) a computationally light, generally applicable, first-principles technique for calculating neutral excitations based on generalized constrained DFT.
Project Title Simulation of dielectric anisotropy at Si surfaces and interfaces
Project Code HPC_18_01023
Principal Investigator Dr Charles Patterson
Start Date 2018-06-11
End Date 2018-08-17
Abstract To continue first principles theoretical work in collaboration with the group of Prof. Thomas Hannappel in Ilmenau, Germany who use dielectric anisotropy to determine properties of buried semiconductor interfaces.
Project Title Simulation of dielectric anisotropy at Si surfaces and interfaces
Project Code HPC_18_01022
Principal Investigator Dr Charles Patterson
Start Date 2018-06-11
End Date 2018-08-17
Abstract To continue first principles theoretical work in collaboration with the group of Prof. Thomas Hannappel in Ilmenau, Germany who use dielectric anisotropy to determine properties of buried semiconductor interfaces.
Project Title Thermodynamic Stability of Heusler Alloys
Project Code HPC_18_01020
Principal Investigator Prof Stefano Sanvito
Start Date 2018-05-08
End Date 2018-08-15
Abstract This project investigates the thermodynamic stability of magnetic Heusler alloys. We'll be looking at different crystal structures, magnetic configurations of selected Heusler alloys.
Project Title 3D models of stellar and exoplanetary outflows
Project Code HPC_18_01019
Principal Investigator Prof Aline Vidotto
Start Date 2018-04-03
End Date 2019-04-03
Abstract We propose to use the computational resources here to model the outflows from low-mass stars and exoplanets. The interaction between stars and exoplanets is an emerging field from the study of stellar winds. With the era of exoplanetary discovery, the understanding of how individual planets interact with their host stars has become more of a topic of discussion. Yet, the field of stellar winds is not yet fully understood, we can only observe certain physical parameters of these stars, such as rotation rate, and surface magnetic fields, but direct observation of low-mass stellar winds are much more difficult due to their tenuous nature. Therefore we must use simulations of stellar winds to infer processes that are occurring within these stellar atmospheres using the data we can from observations as initial and boundary conditions. From these simulations, we can calculate important stellar parameters such as mass-loss rate and angular momentum-loss rate, as well as the magnetic and physical topology of the wind. These parameters allow the community to advance our empirical and theoretical models of these stars and how they evolve. There have now been almost 4,000 exoplanets discovered around other stars. Modelling the interaction of these exoplanets with their host star's wind will give important insights into the conditions surrounding the exoplanets, such as plasma conditions (temperature, velocity, density), bow shock formation, star-planet interactions, and planetary atmospheric loss. This information will guide the next generation of observations on where to look for signatures of these exoplanets. Thereby discovering more exoplanets and increasing our understanding of exoplanetary formation and processes.
Project Title A BLOCK RECYCLED GMRES METHOD WITH INVESTIGATIONS INTO ASPECTS OF SOLVER PERFORMANCE
Project Code HPC_18_01018
Principal Investigator Dr Kirk Soodhalter
Start Date 2018-03-30
End Date 2018-09-30
Abstract We present a block Krylov subspace version of the GCRO-DR method proposed in [Parks et al., SISC 2005]. This new iterative method allows for the efficient minimization of the residual over an augmented Krylov subspace. We offer a new derivation of our proposed method and discuss techniques for selecting recycling subspaces at restart as well as implementation decisions in the context of high-performance computing. Two kinds of numerical experiments are presented: those demonstrating convergence properties, and those demonstrating the data movement and cache efficiencies of the dominant operations of the method, measured using processor monitoring code from Intel.