Parallel dynamic load-balancing for adaptive unstructured meshes

This chapter describes a parallel optimization technique that incorporates a distributed load-balancing algorithm and provides an extremely fast solution to the problem of load-balancing adaptive unstructured meshes. Moreover, a parallel graph contraction technique can be employed to enhance the partition quality and the resulting strategy outperforms or matches results from existing state-of-the-art static mesh partitioning algorithms. The strategy can also be applied to static partitioning problems. Dynamic procedures have been found to be much faster than static techniques, to provide partitions of similar or higher quality and, in comparison, involve the migration of a fraction of the data. The method employs a new iterative optimization technique that balances the workload and attempts to minimize the interprocessor communications overhead. Experiments on a series of adaptively refined meshes indicate that the algorithm provides partitions of an equivalent or higher quality to static partitioners (which do not reuse the existing partition) and much more quickly. The dynamic evolution of load has three major influences on possible partitioning techniques; cost, reuse, and parallelism. The unstructured mesh may be modified every few time-steps and so the load-balancing must have a low cost relative to that of the solution algorithm in between remeshing.

Dynamic load-balancing for parallel adaptive unstructured meshes

A parallel method for dynamic partitioning of unstructured meshes is described. The method employs a new iterative optimisation technique which both balances the workload and attempts to minimise the interprocessor communications overhead. Experiments on a series of adaptively refined meshes indicate that the algorithm provides partitions of an equivalent or higher quality to static partitioners (which do not reuse the existing partition) and much more quickly. Perhaps more importantly, the algorithm results in only a small fraction of the amount of data migration compared to the static partitioners.

Load balancing for parallel adaptive FEM

We present a dynamic distributed load balancing algorithm for parallel, adaptive finite element simulations using preconditioned conjugate gradient solvers based on domain-decomposition. The load balancer is designed to maintain good partition aspect ratios. It calculates a balancing flow using different versions of diffusion and a variant of breadth first search. Elements to be migrated are chosen according to a cost function aiming at the optimization of subdomain shapes. We show how to use information from the second step to guide the first. Experimental results using Bramble's preconditioner and comparisons to existing state-of-the-art balancers show the benefits of the construction.

Dynamic load balancing of distributed memory parallel computational mechanics using unstructured meshes for multi-physical modelling

As the complexity of parallel applications increase, the performance limitations resulting from computational load imbalance become dominant. Mapping the problem space to the processors in a parallel machine in a manner that balances the workload of each processors will typically reduce the run-time. In many cases the computation time required for a given calculation cannot be predetermined even at run-time and so static partition of the problem returns poor performance. For problems in which the computational load across the discretisation is dynamic and inhomogeneous, for example multi-physics problems involving fluid and solid mechanics with phase changes, the workload for a static subdomain will change over the course of a computation and cannot be estimated beforehand. For such applications the mapping of loads to process is required to change dynamically, at run-time in order to maintain reasonable efficiency. The issue of dynamic load balancing are examined in the context of PHYSICA, a three dimensional unstructured mesh multi-physics continuum mechanics computational modelling code.

Mesh partitioning and load balancing for distributed memory parallel systems

A method is outlined for optimising graph partitions which arise in mapping unstructured mesh calculations to parallel computers. The method employs a relative gain iterative technique to both evenly balance the workload and minimise the number and volume of interprocessor communications. A parallel graph reduction technique is also briefly described and can be used to give a global perspective to the optimisation. The algorithms work efficiently in parallel as well as sequentially and when combined with a fast direct partitioning technique (such as the Greedy algorithm) to give an initial partition, the resulting two-stage process proves itself to be both a powerful and flexible solution to the static graph-partitioning problem. Experiments indicate that the resulting parallel code can provide high quality partitions, independent of the initial partition, within a few seconds. The algorithms can also be used for dynamic load-balancing, reusing existing partitions and in this case the procedures are much faster than static techniques, provide partitions of similar or higher quality and, in comparison, involve the migration of a fraction of the data.

Automatic implementation of dynamic load balancing strategies for structured computational mechanics codes

This paper presents a new dynamic load balancing technique for structured mesh computational mechanics codes in which the processor partition range limits of just one of the partitioned dimensions uses non-coincidental limits, as opposed to using coincidental limits in all of the partitioned dimensions. The partition range limits are 'staggered', allowing greater flexibility in obtaining a balanced load distribution in comparison to when the limits are changed 'globally'. as the load increase/decrease on one processor no longer restricts the load decrease/increase on a neighbouring processor. The automatic implementation of this 'staggered' load balancing strategy within an existing parallel code is presented in this paper, along with some preliminary results.

Load balancing for constraint solving with GPUs

Solving a complex Constraint Satisfaction Problem (CSP) is a computationally hard task which may require a considerable amount of time. Parallelism has been applied successfully to the job and there are already many applications capable of harnessing the parallel power of modern CPUs to speed up the solving process. Current Graphics Processing Units (GPUs), containing from a few hundred to a few thousand cores, possess a level of parallelism that surpasses that of CPUs and there are much less applications capable of solving CSPs on GPUs, leaving space for further improvement. This paper describes work in progress in the solving of CSPs on GPUs, CPUs and other devices, such as Intel Many Integrated Cores (MICs), in parallel. It presents the gains obtained when applying more devices to solve some problems and the main challenges that must be faced when using devices with as different architectures as CPUs and GPUs, with a greater focus on how to effectively achieve good load balancing between such heterogeneous devices.

Load effect in abrasive wear mechanism of cast iron with graphite and cementite

In this study four irons were casted with different chromium and vanadium contents: 2.66% Cr, 5.01% Cr, 2.51% V and 5.19% V. Their microstructure is composed of: ledeburite, graphite and M(3)C carbides (cementite). Pin-abrasion tests were carried out using fixed alumina abrasive grains at different loads: 1, 2, 4.6 and 10 N. The wear surface and the abrasive paper were examined by scanning electron microscopy for identifying the wear micromechanism. The results reveal that the mass loss increased with the load increase, and the effect of the percentage of chromium on mass loss is inverted when the load is increased from 4.6 to 10 N; for 4.6 N the mass loss decreased when the chromium percentage was increased from 2.66% to 5.01%. Nevertheless, for 10 N the mass loss increased when the chromium percentage was increased. The worn surfaces of the materials tested at 1 N show microcutting caused by the abrasive tip that produces continuous microchips. The worn surfaces and the abrasive paper tested at 10 N show continuous microchips and brittle debris. The results show that high pressures produce a brittle wear mechanism and low pressures produce a more ductile wear micromechanism, for this, the applied pressure defines the dependence between the wear resistance and wear micromechanism. (C) 2009 Elsevier B.V. All rights reserved.

The need for standardized tests to evaluate the reliability of data transport in wireless medical systems

Wireless medical systems are comprised of four stages, namely the medical device, the data transport, the data collection and the data evaluation stages. Whereas the performance of the first stage is highly regulated, the others are not. This paper concentrates on the data transport stage and argues that it is necessary to establish standardized tests to be used by medical device manufacturers to provide comparable results concerning the communication performance of the wireless networks used to transport medical data. Besides, it suggests test parameters and procedures to be used to produce comparable communication performance results.

Typical load profiles in the smart grid context – a clustering methods comparison

The present research paper presents five different clustering methods to identify typical load profiles of medium voltage (MV) electricity consumers. These methods are intended to be used in a smart grid environment to extract useful knowledge about customer’s behaviour. The obtained knowledge can be used to support a decision tool, not only for utilities but also for consumers. Load profiles can be used by the utilities to identify the aspects that cause system load peaks and enable the development of specific contracts with their customers. The framework presented throughout the paper consists in several steps, namely the pre-processing data phase, clustering algorithms application and the evaluation of the quality of the partition, which is supported by cluster validity indices. The process ends with the analysis of the discovered knowledge. To validate the proposed framework, a case study with a real database of 208 MV consumers is used.

Urubu: energy scavenging in wireless sensor networks

For the past years wireless sensor networks (WSNs) have been coined as one of the most promising technologies for supporting a wide range of applications. However, outside the research community, few are the people who know what they are and what they can offer. Even fewer are the ones that have seen these networks used in real world applications. The main obstacle for the proliferation of these networks is energy, or the lack of it. Even though renewable energy sources are always present in the networks environment, designing devices that can efficiently scavenge that energy in order to sustain the operation of these networks is still an open challenge. Energy scavenging, along with energy efficiency and energy conservation, are the current available means to sustain the operation of these networks, and can all be framed within the broader concept of “Energetic Sustainability”. A comprehensive study of the several issues related to the energetic sustainability of WSNs is presented in this thesis, with a special focus in today’s applicable energy harvesting techniques and devices, and in the energy consumption of commercially available WSN hardware platforms. This work allows the understanding of the different energy concepts involving WSNs and the evaluation of the presented energy harvesting techniques for sustaining wireless sensor nodes. This survey is supported by a novel experimental analysis of the energy consumption of the most widespread commercially available WSN hardware platforms.

Workload balancing in distributed virtual reality environments

Virtual Reality (VR) has grown to become state-of-theart technology in many business- and consumer oriented E-Commerce applications. One of the major design challenges of VR environments is the placement of the rendering process. The rendering process converts the abstract description of a scene as contained in an object database to an image. This process is usually done at the client side like in VRML [1] a technology that requires the client’s computational power for smooth rendering. The vision of VR is also strongly connected to the issue of Quality of Service (QoS) as the perceived realism is subject to an interactive frame rate ranging from 10 to 30 frames-per-second (fps), real-time feedback mechanisms and realistic image quality. These requirements overwhelm traditional home computers or even high sophisticated graphical workstations over their limits. Our work therefore introduces an approach for a distributed rendering architecture that gracefully balances the workload between the client and a clusterbased server. We believe that a distributed rendering approach as described in this paper has three major benefits: It reduces the clients workload, it decreases the network traffic and it allows to re-use already rendered scenes.

RadiaLE: A framework for designing and assessing link quality estimators in wireless sensor networks

Stringent cost and energy constraints impose the use of low-cost and low-power radio transceivers in large-scale wireless sensor networks (WSNs). This fact, together with the harsh characteristics of the physical environment, requires a rigorous WSN design. Mechanisms for WSN deployment and topology control, MAC and routing, resource and mobility management, greatly depend on reliable link quality estimators (LQEs). This paper describes the RadiaLE framework, which enables the experimental assessment, design and optimization of LQEs. RadiaLE comprises (i) the hardware components of the WSN testbed and (ii) a software tool for setting-up and controlling the experiments, automating link measurements gathering through packets-statistics collection, and analyzing the collected data, allowing for LQEs evaluation. We also propose a methodology that allows (i) to properly set different types of links and different types of traffic, (ii) to collect rich link measurements, and (iii) to validate LQEs using a holistic and unified approach. To demonstrate the validity and usefulness of RadiaLE, we present two case studies: the characterization of low-power links and a comparison between six representative LQEs. We also extend the second study for evaluating the accuracy of the TOSSIM 2 channel model.

Radio link quality estimation in wireless sensor networks: a survey

Radio link quality estimation in Wireless Sensor Networks (WSNs) has a fundamental impact on the network performance and also affects the design of higher-layer protocols. Therefore, for about a decade, it has been attracting a vast array of research works. Reported works on link quality estimation are typically based on different assumptions, consider different scenarios, and provide radically different (and sometimes contradictory) results. This article provides a comprehensive survey on related literature, covering the characteristics of low-power links, the fundamental concepts of link quality estimation in WSNs, a taxonomy of existing link quality estimators, and their performance analysis. To the best of our knowledge, this is the first survey tackling in detail link quality estimation in WSNs. We believe our efforts will serve as a reference to orient researchers and system designers in this area.