Peer-to-peer solution of 2D cutting stocks problems

Didier El Baz (1) , Mhand Hifi (2, 3) , Toufik Saadi (4)

1 LAAS-CDA - Équipe Calcul Distribué et Asynchronisme
2 CES - Centre d'économie de la Sorbonne
3 MIS - Modélisation, Information et Systèmes - UR UPJV 4290
4 EPROAD - Eco-Procédés Optimisation et Aide à la Décision - UR UPJV 4669

Peer-to-peer (P2P) applications have known great developments these years. These applications were originally designed for file sharing and are now considered to a larger scope from video streaming to system update and distributed data base. Recent advances in microprocessors architecture and networks permit one to consider new applications like High Performance Computing (HPC). In this paper, we study the parallel solution of 2D cutting stock problems with the peer-to-peer P2PDC environment. First, we present briebly the decentralized version of P2PDC. Then, we propose a distributed P2P algorithm based on dynamic programming and beam search. Finally, we display and analyze computational results.

Mots clés: parallel computing; distributed computing; peer-to-peer com-puting; 2D cutting stock problems; beam search; algorithm

Domaines: Calcul parallèle, distribué et partagé [cs.DC]; Mathématique discrète [cs.DM]

Didier El Baz, Mhand Hifi, Toufik Saadi. Peer-to-peer solution of 2D cutting stocks problems. 11th Cologne - Twente International Workshop on on Graphs and Combinatorial Optimization, May 2012, Munich, Germany. pp.116-120. ⟨hal-01152353⟩