Joint QoS optimization for layered computational grid

Many existing grid resource allocation and scheduling algorithms mainly focus on isolated layers of the grid architecture. The inflexibility of the strict layering structure results in an inefficient utilization of the grid resources. This paper takes a system view of the computational grid and aims to jointly optimize global QoS by adopting cross-layer design. Cross-layer design is based on information exchange and joint optimization among multiple grid layers. Parameters from different layers are provided to a cross-layer optimizer, which selects the values of the layer specific parameters maximizing joint global QoS. The objective of the paper is to jointly optimize the parameters of all layers in a decentralized optimization problem and decompose joint QoS optimization into three sub problems at fabric layer, collective layer and application layer. In simulation part, we compare the performance of the global joint QoS optimization approach with application layer local optimization and resource layer local optimization approach, respectively.     

Cross-layer optimization policy for QoS scheduling in computational grid

This paper presents a cross-layer quality of service (QoS) optimization policy for computational grid. Efficient QoS management is critical for computational grid to meet heterogeneity and dynamics of resources and users’ requirements. There are different QoS metrics at different layers of computational grid. To improve perceived QoS by end users over computational grid, QoS supports can be addressed in different layers, including application layer, collective layer, fabric layer and so forth. The paper tackles cross-layer grid QoS optimization as optimization decomposition, each layer corresponds to a decomposed subproblem. The proposed policy produces an optimal set of grid resources, service compositions and user's payments at the fabric layer, collective layer and application layer respectively to maximize global grid QoS. The cross-layer optimization problem decomposes into three subproblems: grid resource allocation problem, service composing and user satisfaction degree maximization problem, all of which interact through the optimal variables for capacities of grid resources and service demand. In order to coordinate the subproblems, cross-layer QoS feedback mechanism is established to ensure different layer interactions. The simulations are conducted to validate the efficiency of the proposed policy.     

Multi-level scheduling for global optimization in grid computing

The paper presents a multi-level scheduling algorithm for global optimization in grid computing. This algorithm provides a global optimization through a cross-layer optimization realized by decomposing the optimization problem in different sub-problems each of them corresponding to one among the grid layers such as application layer, collective layer and fabric layer. The QoS of an abstraction level is a utility function that assigns at every level a different value and that depends on the kind of task that is executed on the grid. The global QoS is given by processing of the utility function values of the three different levels, using the Lagrangian method. Multi-level QoS scheduling algorithm is evaluated in terms of system efficiency and their economic efficiency, respectively. Economic efficiency includes user utility, service provider’s revenue and grid global utility. System efficiency includes execution success ratio and resource allocation ratio.     

Non-iterative approach for global mesh optimization

This paper presents a global optimization operator for arbitrary meshes. The global optimization operator is composed of two main terms, one part is the global Laplacian operator of the mesh which keeps the fairness and another is the constraint condition which reserves the fidelity to the mesh. The global optimization operator is formulated as a quadratic optimization problem, which is easily solved by solving a sparse linear system. Our global mesh optimization approach can be effectively used in at least three applications: smoothing the noisy mesh, improving the simplified mesh, and geometric modeling with subdivision-connectivity. Many experimental results are presented to show the applicability and flexibility of the approach.     

A hybrid approach to constrained global optimization

In this paper, we propose a novel hybrid global optimization method to solve constrained optimization problems. An exact penalty function is first applied to approximate the original constrained optimization problem by a sequence of optimization problems with bound constraints. To solve each of these box constrained optimization problems, two hybrid methods are introduced, where two different strategies are used to combine limited memory BFGS (L-BFGS) with Greedy Diffusion Search (GDS). The convergence issue of the two hybrid methods is addressed. To evaluate the effectiveness of the proposed algorithm, 18 box constrained and 4 general constrained problems from the literature are tested. Numerical results obtained show that our proposed hybrid algorithm is more effective in obtaining more accurate solutions than those compared to.     

A distributed multiple dimensional QoS constrained resource scheduling optimization policy in computational grid

This paper is to solve efficient QoS based resource scheduling in computational grid. It defines a set of QoS dimensions with utility function for each dimensions, uses a market model for distributed optimization to maximize the global utility. The user specifies its requirement by a utility function. A utility function can be specified for each QoS dimension. In the grid, grid task agent acted as consumer pay for the grid resource and resource providers get profits from task agents. The task agent' utility can then be defined as a weighted sum of single-dimensional QoS utility function. QoS based grid resource scheduling optimization is decomposed to two subproblems: joint optimization of resource user and resource provider in grid market. An iterative multiple QoS scheduling algorithm that is used to perform optimal multiple QoS based resource scheduling. The grid users propose payment for the resource providers, while the resource providers set a price for each resource. The experiments show that optimal QoS based resource scheduling involves less overhead and leads to more efficient resource allocation than no optimal resource allocation.     

A hybrid genetic algorithm and bacterial foraging approach for global optimization

The social foraging behavior of Escherichia coli bacteria has been used to solve optimization problems. This paper proposes a hybrid approach involving genetic algorithms (GA) and bacterial foraging (BF) algorithms for function optimization problems. We first illustrate the proposed method using four test functions and the performance of the algorithm is studied with an emphasis on mutation, crossover, variation of step sizes, chemotactic steps, and the lifetime of the bacteria. The proposed algorithm is then used to tune a PID controller of an automatic voltage regulator (AVR). Simulation results clearly illustrate that the proposed approach is very efficient and could easily be extended for other global optimization problems.     

A stochastic connectionist approach for global optimization withapplication to pattern clustering

In this paper, a stochastic connectionist approach is proposed for solving function optimization problems with real-valued parameters. With the assumption of increased processing capability of a node in the connectionist network, we show how a broader class of problems can be solved. As the proposed approach is a stochastic search technique, it avoids getting stuck in local optima. Robustness of the approach is demonstrated on several multi-modal functions with different numbers of variables. Optimization of a well-known partitional clustering criterion, the squared-error criterion (SEC), is formulated as a function optimization problem and is solved using the proposed approach. This approach is used to cluster selected data sets and the results obtained are compared with that of the K-means algorithm and a simulated annealing (SA) approach. The amenability of the connectionist approach to parallelization enables effective use of parallel hardware.     

Variational optimization for global climate analysis on ESA's high performance computing grid

State of the arte assimilation techniques, such as 3D-Var, are relatively seldom used within climate analysis frameworks, partly because of the enormous numerical costs. In order to face this issue ESA's high performance computing Grid on-Demand (G-POD) is used. We assimilate Global Navigation Satellite System (GNSS) based radio occultations (RO). RO data in general exhibit some favorable properties, like global coverage, all-weather capability expected long-term stability and accuracy. These properties and the continuity of data offered by the Meteorological Operational Satellite (MetOp) program and other RO missions are an ideal opportunity to study the long term atmospheric and climate variability.This paper investigates the assimilation of RO refractivity profiles into first guess fields derived from 21 years of ECMWF's ERA40 dataset on a monthly mean basis divided into four synoptic time layers in order to take the diurnal cycle into account. In contrast to NWP systems, the assimilation procedure is applied without cycling, thus enabling us to run our 3D-Var implementation within G-POD parallel for different time layers. Results indicate a significant analysis increment which is partly systematic, emphasizing the ability of RO data to add independent information to ECMWF analysis fields, with a potential to correct biases. This work lays the ground for further studies using data from existing instruments within a framework based on a mature methodology.     

A New Subdivision Algorithm for the Bernstein Polynomial Approach to Global Optimization

In this paper,an improved algorithm is proposed for unconstrained global optimization to tackle non-convex nonlinear multivariate polynomial programming problems.The proposed algorithm is based on the Bernstein polynomial approach.Novel features of the proposed algorithm are that it uses a new rule for the selection of the subdivision point,modified rules for the selection of the subdivision direction,and a new acceleration device to avoid some unnecessary subdivisions.The performance of the proposed algorithm is numerically tested on a collection of 16 test problems.The results of the tests show the proposed algorithm to be superior to the existing Bernstein algorithm in terms of the chosen performance metrics.     

面向QoS全局优化的大规模Web服务组合方法

把多个简单Web服务组合成为更强大的组合Web服务是面向服务计算的目标之一。由于存在多个功能相同但服务质量属性不同的候选Web 服务,因此需要针对服务质量要求进行服务组合。鉴于Web服务组合规模的不断增长和特定领域的时限要求,面向实时大规模Web服务组合问题的快速收敛算法尤为重要,然而目前相关工作还很少。论文提出一种新的Web服务组合算法GAELS（Genetic Algorithm Embedded Local Searching）,运用高适应度初始种群和局部搜索的变异策略,加快收敛速度。通过实验评测表明与简单遗传算法相比,GAELS算法能更快得到近似最优解,且随着服务规模增长,拥有更好的适应性。     

气象计算网格模式预报系统的建立与优化*

为了在气象部门内跨地域共享使用MM5模式系统,基于建成的气象计算网格平台,成功地接入了MM5模式系统。根据用户预报需求,设计了参数化和业务运行方案,建立了华中区域MM5数值预报应用系统。完成模式系统部署运行后,通过使用网格平台的资源调度和工作流引擎等工具手段,优化了后处理运行流程。运行分析表明网格平台上MM5模式系统的运行效率显著提高。     

Self-adaptive clock synchronization for computational grid

This paper presents an innovative method to synchronize physical clocks for a computational grid, in particular for a computational grid linked through the asynchronous In-tranet or Internet environments. The method discussed is an asynchronous self-adaptive clock synchronization mechanism. Two strategies for clock synchronisation are introduced. (1) Use con-tinuous time intervals to calculate the precision of clocks, which can reduce the effect of network delay effciently. (2) Every node synchronizes its clock with its leader actively. In addition, a node self-adaptive model is presented, and the relationship between the clock precision and synchroniza-tion time is induced, hence a node can predict when it should begin the synchronization process.Detailed simulation and extension of this issue are provided at the end of the paper. The presented model is both practical and feasible.     

A multi-start central force optimization for global optimization

Central force optimization (CFO) is an efficient and powerful population-based intelligence algorithm for optimization problems. CFO is deterministic in nature, unlike the most widely used metaheuristics. CFO, however, is not completely free from the problems of premature convergence. One way to overcome local optimality is to utilize the multi-start strategy. By combining the respective advantages of CFO and the multi-start strategy, a multi-start central force optimization (MCFO) algorithm is proposed in this paper. The performance of the MCFO approach is evaluated on a comprehensive set of benchmark functions. The experimental results demonstrate that MCFO not only saves the computational cost, but also performs better than some state-of-the-art CFO algorithms. MCFO is also compared with representative evolutionary algorithms. The results show that MCFO is highly competitive, achieving promising performance.     

A zonal grid algorithm for DNS of turbulent boundary layers

A zonal grid algorithm for direct numerical simulation (DNS) of incompressible turbulent flows within a Finite-Volume framework is presented. The algorithm uses fully coupled embedded grids and a conservative treatment of the grid-interface variables. A family of conservative prolongation operators is tested in a 2D vortex dipole and a 3D turbulent boundary layer flow. These tests show that both, first- and second-order interpolation conserves the overall second-order spatial accuracy of the scheme. The first-order conservative interpolation has a smaller damping effect on the solution but the second-order conservative interpolation has better spectral properties. The application of this algorithm in boundary layer flow separating and reattaching due to the presence of a streamwise pressure gradient reveals the power and usefulness of the presented algorithm. This simulation has been made possible by the zonal grid algorithm by reducing the required number of grid points from about 500 × 10⁶ to 130 × 10⁶ grid cells.     

A probabilistic algorithm for global optimization

In this paper the authors validate the Wiener process as a suitable stochastic model of the global behaviour of a functionf (x), x∈XCR ¹. An algorithm is given, which exploits the properties of the model, for an effective approximation of the global minimum off (x).     

A decentralized computational algorithm for the global Kalman filter

In this paper a new decentralized computational structure is developed for Optimal state estimation in large scale linear interconnected dynamical systems. The new filter uses a hierarchical structure to perform successive orthogoilalizations on the measurement subspaces of each sub-system in order to provide the optimal estimate. This ensures substantial savings in computation time. In addition, since only low-order subsystem equations are manipulated at each stage, numerical inaccuracies are reduced, and the filter remains stable for even high-order systems. This is illustrated on a multimachine example of a system comprising eleven interconnected machines.     

Parallel resource co-allocation for the computational grid

Resource co-allocation is one of the crucial problems affecting the utility of the grid. Because the numbers of the application tasks and amounts of required resources are enormous and quick responses to the requirements of users are necessary in the real grid environment, real-time resource co-allocation may be large-scale. A parallel resource co-allocation algorithm based on the framework for mapping with resource co-allocation is proposed in this paper. Through the result of experiments, it is concluded that the parallel method reduces the execution time of the resource co-allocation algorithm significantly, and makes the overall response time to the end-users small.     

GCLgrid: A three-dimensional geographical curvilinear grid library for computational seismology

We developed a general library for handling a class of objects we call geographical curvilinear grids (GCLgrids). A GCLgrid is a distorted, uniform grid that is georeferenced. The GCLgrid library is implemented in an object oriented system with methods that relate points in the grid to a geographic reference frame. A simple example is a spherical shell divided along latitude, longitude, and depth, but more elaborate shapes can use the same interface. Scalar and vector fields are derived from the base grid through inheritance. Two algorithms are the core of this library. First, we use the Direction Set method to search for a location in space from a starting point. This algorithm converges reasonably fast if the grid is not extremely distorted. Second, we interpolate the grid using methods known from finite element analysis. A Jacobian matrix for an 8-node cube is computed to transform a distorted cube into a unit one. Shape functions for the standard cube are used to compute interpolation coefficients. Once the interpolation coefficients are computed, we can interpolate n-element vectors almost as quickly as scalar data. We show an application of the library to travel time table calculation at regional distances. Our table interpolator was found to be 10 times faster than one based on the tau-p method and is expected to be several orders of magnitude faster than 3D ray-tracing methods. Travel time interpolation errors are reduced significantly by tabulating delay times relative to a homogenous reference model instead of absolute times. This allows much coarser grids to be used at large scales comparing to one using total time.