HPC

High Performance Computing (HPC). There are at least two ways of interpreting HPC: either as the computational hardware that is used, or as the process of simulating things using a lot of computational power.

We use this tag as a general reference to HPC which is useful for events and work that is too generic to be associated with more particular HPC-relevant tags such as MPI, etc.

The main computational HPC resource at the University of Southampton is Iridis 3 as shown in image (Picture Ash Bennette).

For queries about this topic, contact Hans Fangohr.

View the calendar of events relating to this topic.

Projects

Advanced modelling for two-phase reacting flow

Edward Richardson (Investigator)

Engine designers want computer programs to help them invent ways to use less fuel and produce less pollution. This research aims to provide an accurate and practical model for the injection and combustion of liquid fuel blends.

Aerofoil noise

Richard Sandberg (Investigator)

High-performance computing is used to identify noise sources on aerofoils.

B-meson coupling with relativistic heavy quarks

Jonathan Flynn (Investigator), Patrick Fritzsch, Dirk Broemmel

We non-perturbatively compute the coupling between B* and B pi meson states relying on relativistic heavy quarks and domain wall light fermions. The coupling is of importance for an effective description of hadronic heavy meson decays.

Cellular Automata Modelling of Membrane Formation and Protocell Evolution

Seth Bullock (Investigator), Stuart Bartlett

We simulate the meso-level behaviour of lipid-like particles in a range of chemical and physical environments. Self-organised protocellular structures can be shown to emerge spontaneously in systems with random, homogeneous initial conditions. We aim to take the current model further and explore more complex chemical scenarios in which a broad range of evolutionary behaviours will be exhibited.

Cosmological evolution of supermassive black holes in the centres of galaxies

Anna Kapinska (Investigator)

Abstract to be added...

Development of a novel Navier-Stokes solver (HiPSTAR)

Richard Sandberg (Investigator)

Development of a highly efficient Navier-Stokes solver for HPC.

Development of wide-ranging functionality in ONETEP

Chris-Kriton Skylaris (Investigator), Jacek Dziedzic

ONETEP is at the cutting edge of developments in first principles calculations. However, while the fundamental difficulties of performing accurate first-principles calculations with linear-scaling cost have been solved, only a small core of functionality is currently available in ONETEP which prevents its wide application. In this collaborative project between three Universities, the original developers of ONETEP will lead an ambitious workplan whereby the functionality of the code will be rapidly and significantly enriched.

Fluid Loads and Motions of Damaged Ships

Dominic Hudson, Ming-yi Tan (Investigators), Christian Wood, James Underwood, Adam Sobey

An area of research currently of interest in the marine industry is the effect of damage on ship structures. Research into the behaviour of damaged ships began in the mid nineties as a result of Ro-Ro disasters (e.g. Estonia in 1994). Due to the way the Estonia sank early research mainly focused on transient behaviour immediately after the damage takes place, the prediction of capsize, and of large lateral motions. Further research efforts, headed by the UK MoD, began following an incident where HMS Nottingham ran aground tearing a 50m hole from bow to bridge, flooding five compartments and almost causing the ship to sink just off Lord Howe Island in 2002. This project intends to answer the following questions:
“For a given amount of underwater damage (e.g. collision or torpedo/mine hit), what will be the progressive damage spread if the ship travels at ‘x’ knots? OR for a given amount of underwater damage, what is the maximum speed at which the ship can travel without causing additional damage?”

Graphical Simulation of Archaeological Environments

Graeme Earl (Investigator)

This project defines an emerging area of interest in physically accurate rendering within the Archaeological Computing Research Group. Sub-projects include analysis of Roman spaces at herculaneum, Neolithic buildings at Catalhoyuk and simulation of a range of artefacts.

Hadronic structure on the computer

Jonathan Flynn (Investigator), Dirk Broemmel, Thomas Rae, Ben Samways

In experiments at the Large Hadron Collider (LHC) at CERN, Geneva, the interactions that occur between the colliding particles (protons in this case) can be factorised into a simple scattering between two constituent particles, called quarks, followed by a hadronisation process, which describes the dynamics of forming the bound proton states. Quarks are particles within the proton that bind to form composite particles (hadrons) such as a proton. The scattering process can be computed relatively easily, but hadronisation is intrinsically non-perturbative and hard to calculate. Lattice QCD (computer simulation of QCD on a discrete space-time lattice) provides our only known first-principles and systematically-improvable method to address problems like hadronisation. This project uses Iridis to extract parton distribution amplitudes which are experimentally inaccessible, but needed to describe the quark structure of hadrons.

How far can we stretch the MARTINI?

Syma Khalid (Investigator), Ric Gillams

To date, coarse-grained lipid models have generally been parameterised to ensure the correct prediction of structural properties of membranes, such as the area per lipid and the bilayer thickness. The work described here explores the extent to which coarse-grained models are able to predict correctly bulk properties of lipids (phase behaviour) as well as the mechanical properties, such as lateral pressure profiles and stored elastic stress in bilayers. Such an evaluation is crucial for understanding the predictive capabilities of coarse-grained models.

Hybrid quantum and classical free energy methods in computational drug optimisation

Jonathan Essex, Chris-Kriton Skylaris (Investigators), Christopher Cave-Ayland

This work is based around the application of thermodynamics and quantum mechanics to the field of computational drug design and optimisation. Through the application of these theories the calculation of the physical properties of drug-like molecules is possible and hence some predictive power for their pharmaceutical activity in vivo can be obtained.

Hybrid RANS/LES methods

Richard Sandberg (Investigator), Markus Weinmann

Novel hybrid RANS/LES methods are developed for more accurate and efficient simulation of flow over complex geometries.

Integrated in silico prediction of protein-protein interaction motifs

Richard Edwards (Investigator), Kieren Lythgow

Many vital protein-protein interactions are mediated by Short Linear Motifs (SLiMs) which are short proteins typically 5-15 amino acids long containing only a few positions crucial to function. This project integrates a number of leading computational techniques to predict novel SLiMs and add crucial detail to protein-protein interaction networks.

Is fine-scale turbulence universal?

Richard Sandberg (Investigator)

Complementary numerical simulations and experiments of various canonical flows will try to answer the question whether fine-scale turbulence is universal.

Jet noise

Richard Sandberg (Investigator), Neil Sandham

Direct numerical simulations are used to investigate jet noise.

Mathematical modelling of plant nutrient uptake

Tiina Roose (Investigator)

In this project I will describe a model of plant water and nutrient uptake and how to translate this model and experimental data from the single root scale to the root branching structure scale.

Modelling micromagnetism at elevated temperature

Hans Fangohr (Investigator), Dmitri Chernyshenko

The project aim is to develop a multiscale multiphysics model of
micromagnetism at elevated temperatures combining finite
element/finite difference modelling with atomistic simulations for
material parameter. The tool will be used to guide the development of the next generation magnetic data storage technology: heat assisted magnetic recording.

Modelling power output and wake effects in tidal stream turbine arrays

William Batten (Investigator), Matthew Harrison, Luke Blunden

The PhD research is regards the investigation of modelling techniques for simplifying turbine simulation so that models of large arrays can be investigated.

Multi-objective design optimisation of coronary stents

Neil Bressloff, Georges Limbert (Investigators), Sanjay Pant

Stents are tubular type scaffolds that are deployed (using an inflatable balloon on a catheter), most commonly to recover the shape of narrowed (diseased) arterial segments. Despite the widespread clinical use of stents in cardiovascular intervention, the presence of such devices can cause adverse responses leading to fatality or to the need for further treatment. The most common unwanted responses of inflammation are in-stent restenosis and thrombosis. Such adverse biological responses in a stented artery are influenced by many factors, including the design of the stent. This project aims at using multi-objective optimisation techniques to find an optimum family of coronary stents which are more resistant to the processes of in-stent restenosis (IR) and stent thrombosis (ST).

Non-Perturbative Renormalisation on the Lattice

Jonathan Flynn (Investigator), Dirk Broemmel, Thomas Rae

In this project we compute renormalisation factors for various physical observables in a non-perturbative lattice framework. Renormalisation hereby arises due to a fundamental scale dependence of the physical processes.

Numerical Elastic Neutron Stars

Ian Hawke, Ian Jones (Investigators), Andrew Penner

We study the gravitational wave forms that radiate from an asymmetric neutron star using an elasto-hydrodynamic model.

Prediction of orifice flow flooding rates through generic orifices

Dominic Hudson, Ming-yi Tan (Investigators), Christian Wood, Adam Sobey

This presearch concentrates on the modelling of compartment flooding rates following the occurrence of damage in a ship's side shell. Typical state of the art flooding models use Torricelli’s formula to calculate flooding rates using a constant co-efficient of discharge (Cd). Based on Bernoulli’s theorem, turbulence and viscosity effects are not included using a Cd independent of damage shape or size. Previous work indicates that this assumption over-simplifies the problem to an extent where the flooding rates used for calculation are in error. This project will use CFD validated by experiment to calculate flooding rates for a large number of cases from which a 'krigged' response surface will be generated. Validity of the subsequent response surface will be interrogated.

Scalability of Energy Efficient Routing Algorithms in WSN

Davide Zilli

This project compares two broad classes of routing algorithms for Wireless Sensor Networks, message flooding and single path, by means of a simulation model. In particular, we want to understand how the two scale in terms of energy efficiency on large networks of sensors.

Self-Force and Black Hole Inspirals

Sam Dolan (Investigator)

We use IRIDIS to compute the self-force acting on a solar-mass black hole orbiting a supermassive black hole.

Supersonic axisymmetric wakes

Richard Sandberg (Investigator)

Direct numerical simulations are used to shed more light on structure formation and evolution in supersonic wakes.

Sustainable domain-specific software generation tools for extremely parallel particle-based simulations

Chris-Kriton Skylaris (Investigator)

A range of particle based methods (PBM) are currently used to simulate materials in chemistry, engineering, physics and biophysics. The 4 types of PBM considered directly in the proposed are molecular dynamics (MD), the ONETEP quantum mechanics-based program, discrete element modelling (DEM), and smoothed particle hydrodynamics (SPH).
The overall research objective is to develop a sustainable tool that will deliver, in the future, cutting edge research applicable to applications ranging from dam engineering to atomistic drug design.