Scientific Programming - Volume 17, issue 3

Authors: Knepley, Matthew G. | Karpeev, Dmitry A.

Article Type: Research Article

Abstract: We have developed a new programming framework, called Sieve, to support parallel numerical partial differential equation(s) (PDE) algorithms operating over distributed meshes. We have also developed a reference implementation of Sieve in C++ as a library of generic algorithms operating on distributed containers conforming to the Sieve interface. Sieve makes instances of the incidence relation, or arrows, the conceptual first-class objects represented in the containers. Further, generic algorithms acting on this arrow container are systematically used to provide natural geometric operations on the topology and also, through duality, on the data. Finally, coverings and duality are used to encode not …only individual meshes, but all types of hierarchies underlying PDE data structures, including multigrid and mesh partitions. In order to demonstrate the usefulness of the framework, we show how the mesh partition data can be represented and manipulated using the same fundamental mechanisms used to represent meshes. We present the complete description of an algorithm to encode a mesh partition and then distribute a mesh, which is independent of the mesh dimension, element shape, or embedding. Moreover, data associated with the mesh can be similarly distributed with exactly the same algorithm. The use of a high level of abstraction within the Sieve leads to several benefits in terms of code reuse, simplicity, and extensibility. We discuss these benefits and compare our approach to other existing mesh libraries. Show more

Keywords: Mesh, topology, Sieve, partitioning, distribution

DOI: 10.3233/SPR-2009-0249

Citation: Scientific Programming, vol. 17, no. 3, pp. 215-230, 2009

Authors: Gil, Yolanda

Article Type: Research Article

Abstract: Scientific computing has entered a new era of scale and sharing with the arrival of cyberinfrastructure facilities for computational experimentation. A key emerging concept is scientific workflows, which provide a declarative representation of complex scientific applications that can be automatically managed and executed in distributed shared resources. In the coming decades, computational experimentation will push the boundaries of current cyberinfrastructure in terms of inter-disciplinary scope and integrative models of scientific phenomena under study. This paper argues that knowledge-rich workflow environments will provide necessary capabilities for that vision by assisting scientists to validate and vet complex analysis processes and by automating …important aspects of scientific exploration and discovery. Show more

Keywords: Workflows, computational experimentation, scientific computing, artificial intelligence, automated discovery, knowledge systems

DOI: 10.3233/SPR-2009-0261

Citation: Scientific Programming, vol. 17, no. 3, pp. 231-246, 2009

Authors: Kindratenko, Volodymyr V. | Myers, Adam D. | Brunner, Robert J.

Article Type: Research Article

Abstract: We present a parallel implementation of an algorithm for calculating the two-point angular correlation function as applied in the field of computational cosmology. The algorithm has been specifically developed for a reconfigurable computer. Our implementation utilizes a microprocessor and two reconfigurable processors on a dual-MAP SRC-6 system. The two reconfigurable processors are used as two application-specific co-processors. Two independent computational kernels are simultaneously executed on the reconfigurable processors while data pre-fetching from disk and initial data pre-processing are executed on the microprocessor. The overall end-to-end algorithm execution speedup achieved by this implementation is over 90× as compared to a sequential …implementation of the algorithm executed on a single 2.8 GHz Intel Xeon microprocessor. Show more

Keywords: Reconfigurable computing, angular correlation function

DOI: 10.3233/SPR-2009-0286

Citation: Scientific Programming, vol. 17, no. 3, pp. 247-259, 2009

Authors: Sahni, Onkar | Carothers, Christopher D. | Shephard, Mark S. | Jansen, Kenneth E.

Article Type: Research Article

Abstract: PHASTA falls under the category of high-performance scientific computation codes designed for solving partial differential equations (PDEs). Its a massively parallel unstructured, implicit solver with particular emphasis on fluid dynamics (CFD) applications. More specifically, PHASTA is a parallel, hierarchic, adaptive, stabilized, transient analysis code that effectively employs advanced anisotropic adaptive algorithms and numerical models of flow physics. In this paper, we first describe the parallelization of PHASTA's core algorithms for an implicit solve, where one of our key assumptions is that on a properly balanced supercomputer with appropriate attributes, PHASTA should continue to strongly scale on high core counts until …the computational workload per core becomes insufficient and inter-processor communications start to dominate. We then present and analyze PHASTA's parallel performance across a variety of current near petascale systems, including IBM BG/L, IBM BG/P, Cray XT3, and custom Opteron based supercluster; this selection of systems with inherently different attributes covers a majority of potential candidates for upcoming petascale systems. On one hand, we achieve near perfect (linear) strong scaling out to 32,768 cores of IBM BG/L; showing that a system with desirable attributes will allow implicit solvers to strongly scale on high core counts (including petascale systems). On the contrary, we find that the relative tipping point for strong scaling fundamentally differs among current supercomputer systems. To understand the loss of scaling observed on a particular system (Opteron based supercluster) we analyze the performance and demonstrate that such a loss can be associated to an unbalance in a system attribute; specifically compute-node operating system (OS). In particular, PHASTA scales well to high core counts (up to 32,768 cores) during an implicit solve on systems with compute nodes using lightweight kernels (for example, IBM BG/L); however, we show that on a system where the compute node OS is more heavy weight (e.g., one with background processes) a loss in strong scaling is observed relatively at much fewer number of cores (4,096 cores). Show more

Keywords: Strong scaling, massively parallel processing, unstructured and implicit methods, OS jitter

DOI: 10.3233/SPR-2009-0287

Citation: Scientific Programming, vol. 17, no. 3, pp. 261-274, 2009

Authors: Bushnell, David H.

Article Type: Book Review

DOI: 10.3233/SPR-2009-0284

Citation: Scientific Programming, vol. 17, no. 3, pp. 275-277, 2009

Authors: Nagle, Dan

Article Type: Book Review

DOI: 10.3233/SPR-2009-0285

Citation: Scientific Programming, vol. 17, no. 3, pp. 279-282, 2009