Results 1  10
of
13
The AppLeS Parameter Sweep Template: UserLevel Middleware for the Grid
, 2000
"... The Computational Grid is a promising platform for the efficient execution of parameter sweep applications over large parameter spaces. To achieve performance on the Grid, such applications must be scheduled so that shared data files are strategically placed to maximize reuse, and so that the applic ..."
Abstract

Cited by 196 (29 self)
 Add to MetaCart
The Computational Grid is a promising platform for the efficient execution of parameter sweep applications over large parameter spaces. To achieve performance on the Grid, such applications must be scheduled so that shared data files are strategically placed to maximize reuse, and so that the application execution can adapt to the deliverable performance potential of target heterogeneous, distributed and shared resources. Parameter sweep applications are an important class of applications and would greatly benefit from the development of Grid middleware that embeds a scheduler for performance and targets Grid resources transparently. In this paper we describe a userlevel Grid middleware project, the AppLeS Parameter Sweep Template (APST), that uses applicationlevel scheduling techniques [1] and various Grid technologies to allow the efficient deployment of parameter sweep applications over the Grid. We discuss...
Adaptive Computing on the Grid Using AppLeS
, 2003
"... Ensembles of distributed, heterogeneous resources, also known as Computational Grids are emerging as critical platforms for highperformance and resourceintensive applications. Such platforms provide the potential for applications to aggregate enormous bandwidth, computational power, memory, second ..."
Abstract

Cited by 108 (7 self)
 Add to MetaCart
Ensembles of distributed, heterogeneous resources, also known as Computational Grids are emerging as critical platforms for highperformance and resourceintensive applications. Such platforms provide the potential for applications to aggregate enormous bandwidth, computational power, memory, secondary storage, and other resources during a single execution. However, achieving this performance potential in dynamic, heterogeneous environments is challenging. Recent experience with distributed applications indicates that adaptivity is fundamental to achieving application performance in dynamic Grid environments. The AppLeS (Application Level Scheduling) project provides a methodology, application software, and software environments for adaptively scheduling and deploying applications in dynamic, heterogeneous, multiuser Grid environments. In this paper, we discuss the AppLeS project and outline our results.
A Study of Deadline Scheduling for ClientServer Systems on the Computational Grid
 Proc. of 10th IEEE International Symposium on High Performance Distributed Computing (HPDC10
, 2001
"... The Computational Grid is a promising platform for the deployment of various highperformance computing applications. A number of projects have addressed the idea of software as a service on the network. These systems usually implement clientserver architectures with many servers running on distrib ..."
Abstract

Cited by 24 (0 self)
 Add to MetaCart
The Computational Grid is a promising platform for the deployment of various highperformance computing applications. A number of projects have addressed the idea of software as a service on the network. These systems usually implement clientserver architectures with many servers running on distributed Grid resources and have commonly been referred to as Networkenabled servers (NES). An important question is that of scheduling in this multiclient multiserver scenario. Note that in this context most requests are computationally intensive as they are generated by highperformance computing applications. The Bricks simulation framework has been developed and extensively used to evaluate scheduling strategies for NES systems. In this paper we first present recent developments and extensions to the Bricks simulation models. We discuss a deadline scheduling strategy that is appropriate for the multiclient multiserver case, and augment it with “Load Correction” and “Fallback ” mechanisms which could improve the performance of the algorithm. We then give Bricks simulation results. The results show that future NES systems should use deadlinescheduling with multiple fallbacks and it is possible to allow users to make a tradeoff between failurerate and cost by adjusting the level of conservatism of deadlinescheduling algorithms.
Distributing MCell Simulations on the Grid
 Int. J. High Perform. Comput. Appl
, 2001
"... The Computational Grid [21] is a promising platform for the deployment of largescale scientific and engineering applications. Parameter Sweep Applications (PSAs) arise in many fields of science and engineering and are structured as sets of “experiments”, each of which is executed with a distinct se ..."
Abstract

Cited by 16 (7 self)
 Add to MetaCart
The Computational Grid [21] is a promising platform for the deployment of largescale scientific and engineering applications. Parameter Sweep Applications (PSAs) arise in many fields of science and engineering and are structured as sets of “experiments”, each of which is executed with a distinct set of parameters. Given that structure, PSAs are particularly well suited to the Grid infrastructure and can be deployed on very large scales. However, deployment is not easy to achieve for the domain scientist given the complexity and multiplicity of the Grid software infrastructure, the heterogeneity of the resources, and the dynamic resource availabilities. It is therefore necessary to provide userlevel middleware that acts as an intermediate layer between the application and the Grid. That middleware must address all deployment, data movements, and scheduling issues to provide the user with a transparent way of running his/her simulation on the Grid. In this paper we focus on such middleware specifically targeted to a biology application: MCell. After describing the application and its structure, we describe desired usage scenarios on the Grid. We identify user requirements, discuss relevant computer science issues and propose suitable solutions given currently available Grid technologies. We then describe a general userlevel middleware project for PSAs, APST, explain how it can be extended in order to accommodate MCell’s specific requirements, and introduce current work in that direction. 1.
An Efficient NewtonGMRES Solver for Aerodynamic Computations
 Proceedings of the 13th AIAA CFD Conference, Snowmass
, 1997
"... An efficient inexactNewtonKrylov algorithm is presented for the computation of steady aerodynamic flows. The algorithm uses preconditioned, restarted GMRES in matrixfree form to solve the linear system arising at each Newton iteration. The preconditioner is formed using an ILU(2) factorization of ..."
Abstract

Cited by 11 (4 self)
 Add to MetaCart
An efficient inexactNewtonKrylov algorithm is presented for the computation of steady aerodynamic flows. The algorithm uses preconditioned, restarted GMRES in matrixfree form to solve the linear system arising at each Newton iteration. The preconditioner is formed using an ILU(2) factorization of an approximate Jacobian matrix after applying the Reverse CuthillMcKee reordering. The algorithm has been successfully applied to a wide range of test cases which include inviscid, laminar, and turbulent aerodynamic flows. In all cases except one, convergence of the residual to 10^12 is achieved with a CPU cost equivalent to fewer than 1200 function evaluations. The sole exception is a low Mach number case where some form of local preconditioning is needed. Several other efficient implicit solvers have been applied to the same test cases, and the matrixfree inexactNewtonGMRES algorithm is seen to be the fastest and most robust of the methods studied. Hence this strategy is an excellent option for flow computations in which memory use is not critical, such as twodimensional applications.
A Fast Solver For The Euler Equations On Unstructured Grids Using A NewtonGmres Method
, 1997
"... This paper discusses the use of two variations on Newton 's method, quasiNewton and fullNewton, for the solution of the Euler equations on unstructured triangular grids. The ILU(n)preconditioned GMRES algorithm is employed in the solution of the Jacobian matrix problem which arises at each iterat ..."
Abstract

Cited by 7 (1 self)
 Add to MetaCart
This paper discusses the use of two variations on Newton 's method, quasiNewton and fullNewton, for the solution of the Euler equations on unstructured triangular grids. The ILU(n)preconditioned GMRES algorithm is employed in the solution of the Jacobian matrix problem which arises at each iteration. In the quasiNewton method, a firstorder approximation to the Jacobian matrix is used with the standard GMRES implementation. Both standard and matrixfree implementations of GMRES are used in the fullNewton scheme, and the latter is shown to be much faster and more efficient. A hybrid scheme is presented which makes use of the strengths of both full and quasiNewton implementations, resulting in very fast convergence to steady state. Finally, optimal preconditioning and reordering strategies are presented. Introduction Compressible flow solvers on unstructured grids have proven to be very efficient for steady inviscid flows. The most promising approaches generally use either the m...
NetworkEnabled Server Systems: Deploying Scientific Simulations on the Grid
 In High Performance Computing Symposium (HPC’01
, 2001
"... The Computational Grid [1] is a promising platform for running large scale scientic applications. It provides a base software infrastructure that allows for the development of \middleware" aimed at deploying applications on Grid resources. The NetworkEnabled Server (NES) paradigm is a good candidate ..."
Abstract

Cited by 7 (4 self)
 Add to MetaCart
The Computational Grid [1] is a promising platform for running large scale scientic applications. It provides a base software infrastructure that allows for the development of \middleware" aimed at deploying applications on Grid resources. The NetworkEnabled Server (NES) paradigm is a good candidate as a viable Grid middleware that oers a simple yet powerful programming model (RPCstyle programming for the Grid). This paradigm is amenable to many largescale applications and especially to scienti c simulations. This paper builds on the experience acquired while building two wellknown NES systems (Ninf [2] and NetSolve [3]). Our goal is to clarify major NES design issues as well as to dene a common set of services and concepts that are necessary for implementing and deploying NES systems on the Computational Grid. This paper also describes current work with scientic and engineering simulations that are enabled by NES systems in the Grid context.
A NewtonKrylov Algorithm for the Euler Equations Using Unstructured Grids,” AIAA Paper
, 2003
"... A NewtonKrylov flow solver is presented for the Euler equations on unstructured grids. The algorithm uses a preconditioned matrixfree GMRES method to solve the linear system that arises at each Newton iteration. The preconditioner is an incomplete lowerupper factorization of an approximation to t ..."
Abstract

Cited by 4 (2 self)
 Add to MetaCart
A NewtonKrylov flow solver is presented for the Euler equations on unstructured grids. The algorithm uses a preconditioned matrixfree GMRES method to solve the linear system that arises at each Newton iteration. The preconditioner is an incomplete lowerupper factorization of an approximation to the Jacobian matrix after applying the reverse CuthillMcKee reordering. The algorithm successfully converges for a wide range of steady two and threedimensional aerodynamic flows. A tenorder reduction of the density residual is obtained in a computing time equivalent to fewer than 520 and 1, 800 residual evaluations for the twodimensional and threedimensional cases, respectively.
Parallel Unsteady Turbopump Simulations For Liquid Rocket Engines
, 2000
"... This paper reports the progress being made towards complete turbopump simulation capability for liquid rocket engines. The Space Shuttle Main Engine (SSME) turbopump impeller is used as a test case for the performance evaluation of the MPI, hybrid MPI/OpenMP, and MLP versions of the INS3D code. T ..."
Abstract

Cited by 4 (3 self)
 Add to MetaCart
This paper reports the progress being made towards complete turbopump simulation capability for liquid rocket engines. The Space Shuttle Main Engine (SSME) turbopump impeller is used as a test case for the performance evaluation of the MPI, hybrid MPI/OpenMP, and MLP versions of the INS3D code. Then, a computational model of a turbopump has been developed for the shuttle upgrade program. Relative motion of the grid system for rotorstator interaction was obtained by employing overset grid techniques. Unsteady computations for SSME turbopump, which contains 101 zones with 31 Million grid points, are carried on Origin 2000 systems at NASA Ames Research Center. The approach taken for these simulations, and the performance of the parallel versions of the code are presented. INTRODUCTION The motivation of this effort is based on two primary elements. First, the entire turbo pump simulation intends to provide a computational framework for the design and analysis for the liquid rocket e...
Comparison of Implicit Multigrid Schemes for ThreeDimensional Incompressible Flows
, 2002
"... To develop a robust and efficient computational flow simulation tool for incompressible flow applications, a number of different implicit multigrid schemes for solving the threedimensional incompressible Navier–Stokes equations are compared in the current study. These schemes consist of a common fu ..."
Abstract

Cited by 3 (1 self)
 Add to MetaCart
To develop a robust and efficient computational flow simulation tool for incompressible flow applications, a number of different implicit multigrid schemes for solving the threedimensional incompressible Navier–Stokes equations are compared in the current study. These schemes consist of a common full approximation storage (FAS) multigrid algorithm implemented in conjunction with three different implicit schemes, which include a modified point Gauss relaxation, a standard Gauss–Seidel line relaxation, and the Beam–Warming alternating direction implicit (ADI) scheme. The flow solver used in the study is based on artificial compressibility and uses a thirdorder upwind difference for the convective terms and a secondorder central difference for the viscous terms. The efficiency of each implicit multigrid scheme is assessed in terms of the computing time required for two laminar flow problems: the entry flow through a 90 ◦ bent square duct, and the steadystate and unsteady flows past a prolate spheroid at incidence with an axis ratio of 4:1. It is found that implementation of Neumann boundary conditions on the coarse grid in terms of the flow variable correction rather than the flow variable itself is essential for obtaining