Leroy A. Drummond

An overview of the Advanced CompuTational Software (ACTS) collection

The ACTS Collection brings together a number of general-purpose computational tools that were developed by independent research projects mostly funded and supported by the U.S. Department of Energy. These tools tackle a number of common computational issues found in many applications, mainly implementation of numerical algorithms, and support for code development, execution, and optimization. In this article, we introduce the numerical tools in the collection and their functionalities, present a model for developing more complex computational applications on top of ACTS tools, and summarize applications that use these tools. Last, we present a vision of the ACTS project for deployment of the ACTS Collection by the computational sciences community.

GPU-based 3D lower tree wavelet video encoder

The 3D-DWT is a mathematical tool of increasing importance in those applications that require an efficient processing of huge amounts of volumetric info. Other applications like professional video editing, video surveillance applications, multi-spectral satellite imaging, HQ video delivery, etc, would rather use 3D-DWT encoders to reconstruct a frame as fast as possible. In this article, we introduce a fast GPU-based encoder which uses 3D-DWT transform and lower trees. Also, we present an exhaustive analysis of the use of GPU memory. Our proposal shows good trade off between R/D, coding delay (as fast as MPEG-2 for High definition) and memory requirements (up to 6 times less memory than x264).

A survey of high-quality computational libraries and their impact in science and engineering applications

Recently, a number of important scientific and engineering problems have been successfully studied and solved by means of computational modeling and simulation. Many of these computational models and simulations benefited from the use of available software tools and libraries to achieve high performance and portability. In this article, we present a reference matrix of the performance of robust, reliable and widely used tools mapped to scientific and engineering applications that use them. We aim at regularly maintaining and disseminating this matrix to the computational science community. This matrix will contain information on state-of-the-art computational tools, their applications and their use.

PyACTS: a high-level framework for fast development of high performance applications

Software reusability has proven to be an effective practice to speed-up the development of complex high-performance scientific and engineering applications. We promote the reuse of high quality software and general purpose libraries through the Advance CompuTational Software (ACTS) Collection. ACTS tools have continued to provide solutions to many of todays computational problems. In addition, ACTS tools have been successfully ported to a variety of computer platforms; therefore tremendously facilitating the porting of applications that rely on ACTS functionalities. In this contribution we discuss a high-level user interface that provides a faster code prototype and user familiarization with ACTS tools. The high-level user interfaces have been built using Python. Here we focus on Python based interfaces to ScaLAPACK, the PyScaLAPACK component of PyACTS. We briefly introduce their use, functionalities, and benefits. We illustrate a few simple example of their use, as well as exemplar utilization inside large scientific applications. We also comment on existing Python interfaces to other ACTS tools. We present some comparative performance results of PyACTS based versus direct LAPACK and ScaLAPACK code implementations.

Introduction to the special issue on the Advanced CompuTational Software (ACTS) collection

The ACTS Collection brings together a number of general-purpose computational tools that were developed in independent research projects supported mostly by the U.S. Department of Energy. These tools address common problems found in many engineering and computational sciences applications. ACTS tools implement numerical algorithms and provide functionality such as the solution of linear equations, eigenvalue problems, nonlinear algebraic equations, nonlinear optimization problems, ordinary differential equations, and differential-algebraic equations. In addition, there are utilities for runtime support, profiling, and code and library development. The tools in the collection are focused especially on the solution of large-scale problems on high-performance parallel computers. Many are already well known, like ScaLAPACK and PETSc. Others are modern versions of mathematical software of excellent lineage. Still others are more recently developed. The ACTS Project itself provides an umbrella under which meta-issues associated with mathematical software can be addressed, such as maintenance, support, and interaction with the user community. As mathematical software and the environments in which it must operate become increasingly complex, such issues are emerging as critical. The purpose of this issue is to provide an overview of the project and some of the principal numerical tools of the ACTS collection. The articles present recent mathematical software research that has led to the development of the tools. The articles should be of interest on different levels. Potential users will get a glimpse at a robust collection of state-of-the-art tools that have been developed and improved by the numerical research community over the past 15 years. Numerical algorithm developers will find many insights into techniques for the construction of efficient, reliable, and usable software tools. The articles published in this issue were solicited from participants in the ACTS project. All manuscripts submitted were subjected to the usual ACM TOMS reviewing and refereeing process. Those that have been accepted appear in this special issue. Several others remain under review and may appear in future issues.

Toward Restarting Strategies Tuning for a Krylov Eigenvalue Solver

rylov eigensolvers are used in many scientific fields, such as nuclear physics, page ranking, oil and gas exploration, etc. In this paper, we focus on the ERAM Krylov eigensolver whose convergence is strongly correlated to the Krylov subspace size and the restarting vector \(v_0\), a unit norm vector. We focus on computing the restarting vector \(v_0\) to accelerate the ERAM convergence. First, we study different restarting strategies and compare their efficiency. Then, we mix these restarting strategies and show the considerable ERAM convergence improvement. Mixing the restarting strategies optimizes the “numerical efficiency” versus “execution time” ratio as we do not introduce neither additionnal computation nor communications.

A Communication Optimization Scheme for Basis Computation of Krylov Subspace Methods on Multi-GPUs

Krylov Subspace Methods (KSMs) are widely used for solving large-scale linear systems and eigenproblems. However, the computation of Krylov subspace bases suffers from the overhead of performing global reduction operations when computing the inner vector products in the orthogonalization steps. In this paper, a hypergraph based communication optimization scheme is applied to Arnoldi and incomplete Arnoldi methods of forming Krylov subspace basis from sparse matrix, and features of these methods are compared in a analytical way. Finally, experiments on a CPU-GPU heterogeneous cluster show that our optimization improves the Arnoldi methods implementations for a generic matrix, and a benefit of up to 10x speedup for some special diagonal structured matrix. The performance advantage also varies for different subspace sizes and matrix formats, which requires a further integration of auto-tuning strategy.

An expanded framework for the advanced computational testing and simulation toolkit

The Advanced Computational Testing and Simulation (ACTS) Toolkit is a set of computational tools developed primarily at DOE laboratories and is aimed at simplifying the solution of common and important computational problems. The use of the tools reduces the development time for new codes and the tools provide functionality that might not otherwise be available. This document outlines an agenda for expanding the scope of the ACTS Project based on lessons learned from current activities. Highlights of this agenda include peer-reviewed certification of new tools; finding tools to solve problems that are not currently addressed by the Toolkit; working in collaboration with other software initiatives and DOE computer facilities; expanding outreach efforts; promoting interoperability, further development of the tools; and improving functionality of the ACTS Information Center, among other tasks. The ultimate goal is to make the ACTS tools more widely used and more effective in solving DOEs and the nations scientific problems through the creation of a reliable software infrastructure for scientific computing.