Implementation of a proactive load sharing scheme

This paper presents a proactive approach to load sharing and describes the architecture of a scheme, Concert, based on this approach. A proactive approach is characterized by a shift of emphasis from reacting to load imbalance to avoiding its occurrence. In contrast, in a reactive load sharing scheme, activity is triggered when a processing node is either overloaded or underloaded. The main drawback of this approach is that a load imbalance is allowed to develop before costly corrective action is taken. Concert is a load sharing scheme for loosely-coupled distributed systems. Under this scheme, load and task behaviour information is collected and cached in advance of when it is needed. Concert uses Linux as a platform for development. Implemented partially in kernel space and partially in user space, it achieves transparency to users and applications whilst keeping the extent of kernel modifications to a minimum. Non-preemptive task transfers are used exclusively, motivated by lower complexity, lower overheads and faster transfers. The goal is to minimize the average response-time of tasks. Concert is compared with other schemes by considering the level of transparency it provides with respect to users, tasks and the underlying operating system.

Load sharing in cluster service provision

The factors that are driving the development and use of grids and grid computing, such as size, dynamic features, distribution and heterogeneity, are also pushing to the forefront service quality issues. These include performance, reliability and security. Although grid middleware can address some of these issues on a wider scale, it has also become imperative to ensure adequate service provision at local level. Load sharing in clusters can contribute to the provision of a high quality service, by exploiting both static and dynamic information. This paper is concerned with the presentation of a load sharing scheme, that can satisfy grid computing requirements. It follows a proactive, non preemptive and distributed approach. Load information is gathered continuously before it is needed, and a task is allocated to the most appropriate node for execution. Performance and reliability are enhanced by the decentralised nature of the scheme and the symmetric roles of the nodes. In addition, the scheme exhibits transparency characteristics that facilitate integration with the grid.

Monotone scheme and boundary conditions for finite volume simulation of magnetohydrodynamic internal flows at high Hartmann number

A monotone scheme for finite volume simulation of magnetohydrodynamic internal flows at high Hartmann number is presented. The numerical stability is analysed with respect to the electromagnetic force. Standard central finite differences applied to finite volumes can only be numerically stable if the vector products involved in this force are computed with a scheme using a fully staggered grid. The electromagnetic quantities (electric currents and electric potential) must be shifted by half the grid size from the mechanical ones (velocity and pressure). An integral treatment of the boundary layers is used in conjunction with boundary conditions for electrically conducting walls. The simulations are performed with inhomogeneous electrical conductivities of the walls and reach high Hartmann numbers in three-dimensional simulations, even though a non-adaptive grid is used.

A generational scheme for partitioning graphs

Graph partitioning divides a graph into several pieces by cutting edges. Very effective heuristic partitioning algorithms have been developed which run in real-time, but it is unknown how good the partitions are since the problem is, in general, NP-complete. This paper reports an evolutionary search algorithm for finding benchmark partitions. Distinctive features are the transmission and modification of whole subdomains (the partitioned units) that act as genes, and the use of a multilevel heuristic algorithm to effect the crossover and mutations. Its effectiveness is demonstrated by improvements on previously established benchmarks.

Domain decomposition using a 2-level correction scheme

The PHYSICA software was developed to enable multiphysics modelling allowing for interaction between Computational Fluid Dynamics (CFD) and Computational Solid Mechanics (CSM) and Computational Aeroacoustics (CAA). PHYSICA uses the finite volume method with 3-D unstructured meshes to enable the modelling of complex geometries. Many engineering applications involve significant computational time which needs to be reduced by means of a faster solution method or parallel and high performance algorithms. It is well known that multigrid methods serve as a fast iterative scheme for linear and nonlinear diffusion problems. This papers attempts to address two major issues of this iterative solver, including parallelisation of multigrid methods and their applications to time dependent multiscale problems.

An explicit scheme for coupling temperature and concentration fields in solidification models

A numerical scheme for coupling temperature and concentration fields in a general solidification model is presented. A key feature of this scheme is an explicit time stepping used in solving the governing thermal and solute conservation equations. This explicit approach results in a local point-by-point coupling scheme for the temperature and concentration and avoids the multi-level iteration required by implicit time stepping schemes. The proposed scheme is validated by predicting the concentration field in a benchmark solidification problem. Results compare well with an available similarity solution. The simplicity of the proposed explicit scheme allows for the incorporation of complex microscale models into a general solidification model. This is demonstrated by investigating the role of dendrite coarsening on the concentration field in the solidification benchmark problem.

A fully polynomial approximation scheme for the single machine weighted total tardiness problem with a common due date

We develop a fully polynomial-time approximation scheme (FPTAS) for minimizing the weighted total tardiness on a single machine, provided that all due dates are equal. The FPTAS is obtained by converting an especially designed pseudopolynomial dynamic programming algorithm.

Virtual assignments: breach of covenant against parting with or sharing of possession

Comments on the Chancery Division decision in Clarence House Ltd v National Westminster Bank Plc on whether the alienation covenant in a lease of commercial premises had been breached by the tenant effecting a virtual assignment of it, under which all the economic benefits and burdens of the lease were transferred to a third party without there being any actually assignment of the leasehold interest or change in occupancy.

Eating secret [at Kingston University]

The conference took place at the Penrhyn Road Campus, between 6 and 8 July 2009. This international conference on 'Performing Lives' followed the inaugural conference of the Centre for Life Narratives, 'The Spirit of the Age', the focus of which was 'Writing Lives' (2007). Our second biennial conference 'Performing Lives' (2009) invited analysis and debate on the relationship between life histories and the ways in which they are embodied and enacted in performance, across a range of cultures and a variety of media: drama, dance, film, TV and video.

Eating secret [at Whitstable Biennale Satellite]

Research around ingesting /public perceptions of ‘secret’ and exploring other narratives around our perceptions of food and eating.

A study on the cost sharing model for supply lead time reduction with component commonality

This paper studies a two-level supply chain consisting of components supplier and product assembly manufacturer, while the manufacturer shares the investment on shortening supply lead time. The objective of this research is to investigate the benefits of cost sharing strategy and adopting component commonality. The result of numerical analysis demonstrates that using component commonality can help reduce the total cost, especially when the manufacture shares a higher fraction of the cost of investment in shortening supply lead time.

Dynamic evolution in information sharing networks

This study examines the roll-out of a collaborative information repository or 'knowledge-base' in a medium-sized UK professional services firm over a six year period. Data from usage logs provides the basis for analysis of the dynamic evolution of social networks around the depository during this time. The adoption pattern follows an 's-curve' and usage exhibits something of a power law distribution, both attributable to network effects and network opposition is associated with organisational performance on a number of indicators. But periodicity in usage is evident and the usage distribution displays an exponential cut-off. Fourier analysis provides some evidence of mathematical complexity in the periodicity. Some implications of complex patterns in social network data for research and management are discussed.