Stingray: Cone Tracing using a software DSM for SCI clusters

Alexandre Meyer 1 Emmanuel Cecchet 2
1 iMAGIS - Models, Algorithms and Geometry for Computer Generated Image Graphics
GRAVIR - IMAG - Graphisme, Vision et Robotique, Inria Grenoble - Rhône-Alpes
Abstract : In this paper we consider the use of a supercomputer with a hardware shared memory versus a cluster of workstations using a software Distributed Shared Mem-ory (DSM). We focus on ray tracing applications to compare both architectures. We have ported Stingray, a parallel cone tracer developed on a SGI Origin 2000 super-computer, on a cluster using a Scalable Coherent Interface (SCI) network and a software DSM called SciFS. We present concepts of cone tracing with Stingray, concepts of SCI cluster with a DSM and the implementa-tion issues. We compare the results obtained with the two architectures and we discuss the trade-off - price/performance/programming ease - of both architectures. We show with Stingray that a modest 12 nodes SCI cluster with an efficient software DSM is 5 times cheaper and can perform up to 2.3 times better than a SGI Origin 2000 with 6 processors. We think that a software DSM is well suited for this kind of applications and provides both ease of programming and scalable per-formance.
Type de document :
Communication dans un congrès
Daniel S. Katz and Thomas Sterling and Mark Baker and Larry Bergman and Marcin Paprzycki and Rajkumar Buyya. IEEE International Conference on Cluster Computing, Oct 2001, Newport Beach, United States. IEEE Computer Society, pp.95 - 101, 2001, 〈http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=959957〉. 〈10.1109/CLUSTR.2001.959957〉
Liste complète des métadonnées

Littérature citée [9 références]  Voir  Masquer  Télécharger


https://hal.inria.fr/inria-00537505
Contributeur : Team Artis <>
Soumis le : jeudi 18 novembre 2010 - 15:49:21
Dernière modification le : jeudi 11 janvier 2018 - 06:20:04
Document(s) archivé(s) le : samedi 19 février 2011 - 03:13:06

Fichiers

MeyerCecchet_Stingray.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

INRIA | UGA | IMAG

Citation

Alexandre Meyer, Emmanuel Cecchet. Stingray: Cone Tracing using a software DSM for SCI clusters. Daniel S. Katz and Thomas Sterling and Mark Baker and Larry Bergman and Marcin Paprzycki and Rajkumar Buyya. IEEE International Conference on Cluster Computing, Oct 2001, Newport Beach, United States. IEEE Computer Society, pp.95 - 101, 2001, 〈http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=959957〉. 〈10.1109/CLUSTR.2001.959957〉. 〈inria-00537505〉

Partager

Métriques

Consultations de la notice

226

Téléchargements de fichiers

199