Efficient Serial and Parallel Coordinate Descent Methods for Huge-Scale Truss Topology Design

Abstract

In this work we propose solving huge-scale instances of the truss topology design problem with coordinate descent methods. We develop four efficient codes: serial and parallel implementations of randomized and greedy rules for the selection of the variable (potential bar) to be updated in the next iteration. Both serial methods enjoy an O(n/k) iteration complexity guarantee, where n is the number of potential bars and k the iteration counter. Our parallel implementations, written in CUDA and running on a graphical processing unit (GPU), are capable of speedups of up to two orders of magnitude when compared to their serial counterparts. Numerical experiments were performed on instances with up to 30 million potential bars.

DOI: 10.1007/978-3-642-29210-1_5

Extracted Key Phrases

2 Figures and Tables

Showing 1-10 of 14 references

Topology optimization of large-scale trusses using ground structure approach with selective subsets of active bars

  • T Sokół
  • 2011

Truss topology design with integer variables made easy

  • M Kočvara
  • 2010

Coordinate descent optimization for l 1 minimization with application to compressed sensing; a greedy algorithm

  • Y Li, S Osher
  • 2009

Layout optimization of large-scale pin-jointed frames

  • M Gilbert, A Tyas
  • 2003

Layout optimization of structures

  • G Rozvany, M P Bendsøe, U Kirsch
  • 1995

Bridge " truss after 20s, 200s (≈3.34 minutes) and 20,000s (≈5.56 hours) of computation time for algorithms SR, PR, SG and PG (rows in this order) Number of iterations

  • Fig
Showing 1-10 of 21 extracted citations