In this paper, we investigate G-commerce — computational economies for controlling resource allocation in Computational Grid settings. We define hypothetical resource consumers (representing users and Grid-aware applications) and resource producers (representing resource owners who “sell ” their resources to the Grid). We then measure the efficiency of resource allocation under two different market conditions: commodities markets and auctions. We compare both market strategies in terms of price stability, market equilibrium, consumer efficiency, and producer efficiency. Our results indicate that commodities markets are a better choice for controlling Grid resources than previously defined auction strategies. 1

This paper presents the design of the XenoServer Open Platform: a public infrastructure for wide-area computing, capable of hosting tasks that span the full spectrum of distributed programming. The platform integrates resource management, charging and auditing. We emphasize the control-plane aspects of the system, showing how it supports service deployment with a low cost of entry and how it forms a substrate over which other distributed computing platforms can be deployed.

Recent advances in computing and communication have given rise to the computational grid notion. The core of this computing paradigm is the design of a system for drawing compute power from a confederation of geographically dispersed heterogeneous resources, seamlessly and ubiquitously. If high-performance levels are to be achieved, data locality must be identified and managed. In this paper, we consider the affect of server side staging on the behavior of a class of wide area “task farming ” applications. We show that staging improves task throughput mainly through the increased parallelism rather than the reduction in overall turnaround time per task. We derive a model for farming applications with and without server side staging and verify the model through live experiments as well as simulations. 1.

The satisfiability (SAT) problem is a central problem in mathematical logic, computing theory, and artificial intelligence. An instance of SAT is specified by a set of boolean variables and a propositional formula in conjunctive normal form. Given such an instance, the SAT problem asks whether there is a truth assignment to the variables such that the formula is satisfied. It is well known that SAT is in general NP-complete, although several important special cases can be solved in polynomial time. Semidefinite programming (SDP) refers to the class of optimization problems where a linear function of a matrix variable X is maximized (or minimized) subject to linear constraints on the elements of X and the additional constraint that X be positive semidefinite. We are interested in the application of SDP to satisfiability problems, and in particular in how SDP can be used to detect unsatisfiability. In this paper we introduce a new SDP relaxation for the satisfiability problem. This SDP relaxation arises from the recently introduced paradigm of “higher liftings” for constructing semidefinite programming relaxations of discrete optimization problems.

With the proliferation of the Internet comes the possibility of aggregating vast collections of computers into largescale computational platforms. Research driven by this realization has yielded a new software architecture, known as The Computational Grid [16], for building high-performance distributed applications and systems. The vision outlined by its architects is for applications to draw computational “power ” from a distributed pool of resources in

videnced, for example, by the optimization-model statements in appendices to many Interfaces papers. To support this model-building activity, specialized modeling systems have been developed for building, analyzing, and maintaining optimization models. These systems handle models independently of the choice of solver and so can be hooked to a variety OPTIMIZATION March--April2rc 131 of solvers. Conversely, popular solvers work with a selection of modeling systems. The analyst who wants to build an optimization application must consequently sort through quite a tangle of software. This is in contrast to the situation to such areas as statistics or simulation, where the software principally consists of integrated packages of modeling tools and modelanalysis methods. And this is where Internet services enter the picture, as providers of guidance and access to the daunting variety of optimization software. Prelimina5d2 By optimiz

The Computational Grid [12] has been proposed for the implementation of high-performance applications using widely dispersed computational resources. The goal of a Computational Grid is to aggregate ensembles of shared, heterogeneous, and distributed resources (potentially controlled by separate organizations) to provide computational “power ” to an application program. In this paper, we provide a toolkit for the development of globally deployable Grid applications. The toolkit, called EveryWare, enables an application to draw computational power transparently from the Grid. It consists of a portable set of processes and libraries that can be incorporated into an application so that a wide variety of dynamically changing distributed infrastructures and resources can be used together to achieve supercomputer-like performance. We provide our experiences gained while building the EveryWare toolkit prototype and an its use in implementing a large-scale Grid application.

Often the most robust and efficient algorithms for the solution of large-scale problems involving nonlinear PDEs and optimization require the computation of derivative quantities. We examine the use of automatic differentiation (AD) to provide code for computing first and second derivatives in conjunction with two parallel numerical toolkits, the Portable, Extensible Toolkit for Scientific Computing (PETSc) and the Toolkit for Advanced Optimization (TAO). We discuss how the use of mathematical abstractions for vectors and matrices in these libraries facilitates the use of AD to automatically generate derivative codes and present performance data demonstrating the suitability of this approach. 1 Introduction As the complexity of advanced computational science applications has increased, the use of object-oriented software methods for the development of both applications and numerical toolkits has also increased. The migration toward this approach can be attributed in part to ...

PVODE is a high-performance ordinary differential equation solver for the types of initial value problems (IVPs) that arise in largescale computational simulations. Often, one wants to compute sensitivities with respect to certain parameters in the IVP. We discuss the use of automatic differentiation (AD) to compute these sensitivities in the context of PVODE. Results on a simple test problem indicate that the use of AD-generated derivative code can reduce the time to solution over finite difference approximations. 1 Background In complicated, large-scale computational simulations, the governing equations can often be spatially discretized and then numerically solved as a system of ordinary differential equation (ODE) or differential-algebraic equation (DAE) initial-value problems. PVODE [BH99] and IDA [HT99] are powerful, parallel codes for solving these types of ODEs and DAEs, respectively. The codes are written in C and use MPI to achieve parallelism and portability. Typica...

Abstract not found