
Analyzing the Performance of Greedy Maximal Scheduling via Local Pooling and Graph Theory
Zussman, Gil; Seymour, Paul; Zwols, Yori; Birand, Berk; Ries, Bernard; Chudnovsky, Maria (2010), Analyzing the Performance of Greedy Maximal Scheduling via Local Pooling and Graph Theory, INFOCOM, 2010 Proceedings IEEE, IEEE Press Book : New York, p. 1-9
Voir/Ouvrir
Type
Communication / ConférenceDate
2010Titre du colloque
29th IEEE International Conference on Computer CommunicationsDate du colloque
2010-03Ville du colloque
San DiegoPays du colloque
États-UnisTitre de l'ouvrage
INFOCOM, 2010 Proceedings IEEEÉditeur
IEEE Press Book
Ville d’édition
New York
Isbn
978-1-4244-5838-7
Pages
1-9
Identifiant publication
Métadonnées
Afficher la notice complèteRésumé (EN)
The main task in analyzing a switching network design (including circuit-, multirate-, and photonic-switching) is to determine the minimum number of some switching components so that the design is non-blocking in some sense (e.g., stridor wide-sense). We show that, in many cases, this task can be accomplished with a simple two-step strategy: (1) formulate a linear program whose optimum value is a bound for the minimum number we are seeking, and (2) specify a solution to the dual program, whose objective value by weak duality immediately yields a sufficient condition for the design to be non-blocking. We illustrate this technique through a variety of examples, ranging from circuit to multirate to photonic switching, from unicast to f-cast and multicast, and from strict- to wide-sense non-blocking. The switching architectures in the examples are of Clos-type and Banyan-type, which are the two most popular architectural choices for designing non-blocking switching networks. To prove the result in the multirate Clos network case, we formulate a new problem called DYNAMIC WEIGHTED EDGE COLORING which generalizes the DYNAMIC BIN PACKING problem. We then design an algorithm with competitive ratio 5.6355 for the problem. The algorithm is analyzed using the linear programming technique. We also show that no algorithm can have competitive ratio better than 4-O (log n/n) for this problem. New lower- and upper-bounds for multirate wide-sense non-blocking Clos networks follow, improving upon a couple of 10-year-old bounds on the same problem.Mots-clés
Greedy Maximal Scheduling; Graph Theory; Local PoolingPublications associées
Affichage des éléments liés par titre et auteur.
-
Birand, Berk; Chudnovsky, Maria; Ries, Bernard; Seymour, Paul; Zussman, Gil; Zwols, Yori (2012) Article accepté pour publication ou publié
-
Ries, Bernard; Chudnovsky, Maria; Zwols, Yori (2011) Article accepté pour publication ou publié
-
Zwols, Yori; Ries, Bernard; Chudnovsky, Maria (2011) Article accepté pour publication ou publié
-
Dereniowski, Dariusz; Kubiak, W.; Ries, Bernard; Zwols, Yori (2013) Article accepté pour publication ou publié
-
Golumbic, Martin Charles; Ries, Bernard (2013) Article accepté pour publication ou publié