Dynamic load balancing in parallel KD-tree k-means

One among the most influential and popular data mining methods is the k-Means algorithm for cluster analysis. Techniques for improving the efficiency of k-Means have been largely explored in two main directions. The amount of computation can be significantly reduced by adopting geometrical constraints and an efficient data structure, notably a multidimensional binary search tree (KD-Tree). These techniques allow to reduce the number of distance computations the algorithm performs at each iteration. A second direction is parallel processing, where data and computation loads are distributed over many processing nodes. However, little work has been done to provide a parallel formulation of the efficient sequential techniques based on KD-Trees. Such approaches are expected to have an irregular distribution of computation load and can suffer from load imbalance. This issue has so far limited the adoption of these efficient k-Means variants in parallel computing environments. In this work, we provide a parallel formulation of the KD-Tree based k-Means algorithm for distributed memory systems and address its load balancing issue. Three solutions have been developed and tested. Two approaches are based on a static partitioning of the data set and a third solution incorporates a dynamic load balancing policy.

An investigation of pile diameter influence in the bearing capacity on Dynamic Load Test (DLT)

During the construction of five residential buildings in the city of Taubate, State of São Paulo, it was possible to carry out one comprehensive investigation of the behavior of precast concrete piles in clay shales. This paper describes the results of Dynamic Load Tests (DLT's) executed in three piles with different diameters and with the same embedded length. The tests were monitored using the PDA(R) (Pile Driving Analyzer) and the pile top displacement was measured by pencil and paper procedure. From the curves of RMX versus DMX resulted from CASE(R) method, CAPWAPC(R) analyses were made for signals where the maximum mobilized soil resistance was verified. The results were compared with the predicted bearing capacity using the semi-empirical method of Decourt & Quaresma (1978) and Decourt (1982) based on SPT values and the description of the soil profile. Some comments related to the values of quake and damping used for clay shales in the analyses are also presented.

Experience gained and difficulties in performing dynamic load test in composite piles made with steel rails

The results of dynamic instrumentation in two construction sites that used steel rails as pile foundation are presented in this paper. The first dynamic load tests were executed in piles made by single steel rails of the type TR32 and TR37. In the second group of dynamic tests, the piles were made by a composition of two and three steel rails of the same type TR37. The difficulties in placing the sensors, the effect of hammer blow eccentricity and the influence of the non-uniform welding along the pile length are presented in detail and discussed.

3-Phase Rectifier System with very demanding dynamic load: Architecture analysis and control strategy

A distributed power architecture for aerospace application with very restrictive specifications is analyzed. Parameters as volume, weight and losses are analyzed for the considered power architectures. In order to protect the 3 phase generator against high load steps, an intermediate bus (based in a high capacitance) to provide energy to the loads during the high load steps is included. Prototypes of the selected architecture for the rectifier and EMI filter are built and the energy control is validated.

A generic strategy for dynamic load balancing of distributed memory parallel computational mechanics using unstructured meshes

A large class of computational problems are characterised by frequent synchronisation, and computational requirements which change as a function of time. When such a problem is solved on a message passing multiprocessor machine [5], the combination of these characteristics leads to system performance which deteriorate in time. As the communication performance of parallel hardware steadily improves so load balance becomes a dominant factor in obtaining high parallel efficiency. Performance can be improved with periodic redistribution of computational load; however, redistribution can sometimes be very costly. We study the issue of deciding when to invoke a global load re-balancing mechanism. Such a decision policy must actively weigh the costs of remapping against the performance benefits, and should be general enough to apply automatically to a wide range of computations. This paper discusses a generic strategy for Dynamic Load Balancing (DLB) in unstructured mesh computational mechanics applications. The strategy is intended to handle varying levels of load changes throughout the run. The major issues involved in a generic dynamic load balancing scheme will be investigated together with techniques to automate the implementation of a dynamic load balancing mechanism within the Computer Aided Parallelisation Tools (CAPTools) environment, which is a semi-automatic tool for parallelisation of mesh based FORTRAN codes.

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.

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.

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.

Stability boundary analysis of the dynamic voltage restorer in weak systems with dynamic loads

In this contribution, a stability analysis for a dynamic voltage restorer (DVR) connected to a weak ac system containing a dynamic load is presented using continuation techniques and bifurcation theory. The system dynamics are explored through the continuation of periodic solutions of the associated dynamic equations. The switching process in the DVR converter is taken into account to trace the stability regions through a suitable mathematical representation of the DVR converter. The stability regions in the Thevenin equivalent plane are computed. In addition, the stability regions in the control gains space, as well as the contour lines for different Floquet multipliers, are computed. Besides, the DVR converter model employed in this contribution avoids the necessity of developing very complicated iterative map approaches as in the conventional bifurcation analysis of converters. The continuation method and the DVR model can take into account dynamics and nonlinear loads and any network topology since the analysis is carried out directly from the state space equations. The bifurcation approach is shown to be both computationally efficient and robust, since it eliminates the need for numerically critical and long-lasting transient simulations.

Investigation of sensitivity based load ranking with composite load model in power system

Load modeling plays an important role in power system dynamic stability assessment. One of the widely used methods in assessing load model impact on system dynamic response is through parametric sensitivity analysis. Load ranking provides an effective measure of such impact. Traditionally, load ranking is based on either static or dynamic load model alone. In this paper, composite load model based load ranking framework is proposed. It enables comprehensive investigation into load modeling impacts on system stability considering the dynamic interactions between load and system dynamics. The impact of load composition on the overall sensitivity and therefore on ranking of the load is also investigated. Dynamic simulations are performed to further elucidate the results obtained through sensitivity based load ranking approach.

Effect of vertical misalignment of adjacent sleepers on the increase in dynamic loads around rail joints

In spite of the extensive usage of continuous welded rails, a number of rail joints still exist in the track. Although a number of them exist as part of turnouts in the yards where the speed is not of concern, the Insultated Rail Joints (IRJs) that exist in ballasted tracks remain a source of significant impact loading. A portion of the dynamic load generated at the rail joints due to wheel passage is transmitted to the support system which leads to permanent settlements of the ballast layer with subsequent vertical misalignment of the sleepers around the rail joints. The vertical misalignment of the adjacent sleepers forms a source of high frequency dynamic load raisers causing significant maintenance work including localised grinding of railhead around the joint, re-alignment of the sleepers and/or ballast tamping or track component renewals/repairs. These localised maintenance activities often require manual inspections and disruptions to the train traffic loading to significant costs to the rail industry. Whilst a number of studies have modelled the effect of joints as dips, none have specifically attended to the effect of vertical misalignment of the sleepers on the dynamic response of rail joints. This paper presents a coupled finite element track model and rigid body track-vehicle interaction model through which the effects of vertical of sleepers on the increase in dynamic loads around the IRJ are studied. The finite element track model is employed to determine the generated dip from elastic deformations as well as the vertical displacement of sleepers around the joint. These data (dip and vertical misalignments) are then imported into the rigid body vehicle-track interaction model to calculate the dynamic loads.

ELISA: An Estimated Load Information Scheduling Algorithm for Distributed Computing Systems

In this paper, we present a decentralized dynamic load scheduling/balancing algorithm called ELISA (Estimated Load Information Scheduling Algorithm) for general purpose distributed computing systems. ELISA uses estimated state information based upon periodic exchange of exact state information between neighbouring nodes to perform load scheduling. The primary objective of the algorithm is to cut down on the communication and load transfer overheads by minimizing the frequency of status exchange and by restricting the load transfer and status exchange within the buddy set of a processor. It is shown that the resulting algorithm performs almost as well as a perfect information algorithm and is superior to other load balancing schemes based on the random sharing and Ni-Hwang algorithms. A sensitivity analysis to study the effect of various design parameters on the effectiveness of load balancing is also carried out. Finally, the algorithm's performance is tested on large dimensional hypercubes in the presence of time-varying load arrival process and is shown to perform well in comparison to other algorithms. This makes ELISA a viable and implementable load balancing algorithm for use in general purpose distributed computing systems.

Load balancing algorithms for an extended hypercube

Reduction of the execution time of a job through equitable distribution of work load among the processors in a distributed system is the goal of load balancing. Performance of static and dynamic load balancing algorithms for the extended hypercube, is discussed. Threshold algorithms are very well-known algorithms for dynamic load balancing in distributed systems. An extension of the threshold algorithm, called the multilevel threshold algorithm, has been proposed. The hierarchical interconnection network of the extended hypercube is suitable for implementing the proposed algorithm. The new algorithm has been implemented on a transputer-based system and the performance of the algorithm for an extended hypercube is compared with those for mesh and binary hypercube networks

Evaluation of the accuracy of an impact load cell using lumped parameter modeling and analysis

The objective of the current study is to evaluate the fidelity of load cell reading during impact testing in a drop-weight impactor using lumped parameter modeling. For the most common configuration of a moving impactor-load cell system in which dynamic load is transferred from the impactor head to the load cell, a quantitative assessment is made of the possible discrepancy that can result in load cell response. A 3-DOF (degrees-of-freedom) LPM (lumped parameter model) is considered to represent a given impact testing set-up. In this model, a test specimen in the form of a steel hat section similar to front rails of cars is represented by a nonlinear spring while the load cell is assumed to behave in a linear manner due to its high stiffness. Assuming a given load-displacement response obtained in an actual test as the true behavior of the specimen, the numerical solution of the governing differential equations following an implicit time integration scheme is shown to yield an excellent reproduction of the mechanical behavior of the specimen thereby confirming the accuracy of the numerical approach. The spring representing the load cell, however,predicts a response that qualitatively matches the assumed load-displacement response of the test specimen with a perceptibly lower magnitude of load.