Journal Article10.1137/0805029
Data parallel quadratic programming on box-constrained problems*
4
TL;DR: Designs for the data parallel solution of quadratic programming problems subject to box constraints are developed and the class of algorithms that iterate between projection steps that identify candidate active sets and conjugate gradient steps that explore the working space are considered.
read more
Abstract: We develop designs for the data parallel solution of quadratic programming problems subject to box constraints. In particular, we consider the class of algorithms that iterate between projection steps that identify candidate active sets and conjugate gradient steps that explore the working space. Using the algorithm of More and Toraldo [Report MCS-p77-05 89, Argonne National Laboratory, Illinois, 1989] as a specific instance of this class of algorithms we show how its components can be implemented efficiently on a data-parallel SIMD computer architecture. Alternative designs are developed for both arbitrary, unstructured Hessian matrices and for structured problems.Implementations are carried out on a Connection Machine CM–2. They are shown to be very efficient, achieving a peak computing rate over 2 Gflops. Problems with several hundred thousand variables are solved within one minute of solution time on the 8K CM–2. Extremely large test problems, with up to 2.89 million variables, are also solved efficie...
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
•Posted Content
GPCG: A Case Study in the Performance and Scalability of Optimization Algorithms
TL;DR: This case study uses the GPCG (gradient projection, conjugate gradient) algorithm for solving bound-constrained convex quadratic problems and analyzes the performance as a function of the number of variables, thenumber of free variables, and the preconditioner.
A case study in the performance and scalability of optimization algorithms
TL;DR: In this paper, the performance and scalabilty of algorithms for the solution of large optimization problems on high-performance parallel architectures were analyzed using the GPCG (gradient projection, conjugate gradient) algorithm for solving boundconstrained convex quadratic problems.
24
A parallel relaxation method for quadratic programming problems with interval constraints
TL;DR: This paper proposes a highly parallelizable method for solving optimization problems with interval constraints, which may be regarded as an application of the Jacobi method to the dual of the original problems.
19
Random test problems and parallel methods for quadratic programs and quadratic stochastic programs
Xiaojun Chen,Robert S. Womrsley +1 more
TL;DR: A data parallel procedure for randomly generating test problems for two-stage quadratic stochastic programming that allows the user to specify the size of the problem, the condition numbers of the Hessian matrices of the objective functions and the structure of the feasible regions in the first and the second stages.
References
Mixed finite elements in ℝ 3
TL;DR: In this article, the authors present two families of non-conforming finite elements, built on tetrahedrons or on cubes, which are respectively conforming in the spacesH(curl) and H(div).
3.6K
Performance of various computers using standard linear equations software
TL;DR: This report compares the performance of different computer systems in solving dense systems of linear equations, ranging from a CRAY Y-MP to scientific workstations such as the Apollo and Sun to IBM PCs.
On the Goldstein-Levitin-Polyak gradient projection method
Dimitri P. Bertsekas
- 01 Nov 1974
TL;DR: In this paper, the authors consider the convergence of a gradient projection method to a conjugate direction with a quasi-Newton or Newton's method and achieve the attendant superlinear convergence rate.
Projected Newton methods and optimization of multicommodity flows
Dimitri P. Bertsekas,Eli Gafni +1 more
TL;DR: Preliminary computational results suggest that this type of method is capable of yielding highly accurate solutions of nonlinear multicommodity flow problems far more efficiently than any of the methods available at present.