DMCA
An Evaluation of Scalable Application-Level Multicast Built Using Peer-to-Peer Overlays (2003)
Cached
Download Links
- [www.ieee-infocom.org]
- [www.cs.huji.ac.il]
- [www.cs.usask.ca]
- [infocom2003.ieee-infocom.org]
- [www.research.microsoft.com]
- [www.research.microsoft.com]
- [research.microsoft.com]
- [www.cs.rice.edu]
- [research.microsoft.com]
- [research.microsoft.com]
- [www.comp.nus.edu.sg]
- [www.research.microsoft.com]
- [www.research.microsoft.com]
- [research.microsoft.com]
- [research.microsoft.com]
- [research.microsoft.com]
- [research.microsoft.com]
- [www.research.microsoft.com]
- [research.microsoft.com]
- [research.microsoft.com]
- [www.eecs.harvard.edu]
- [research.microsoft.com]
- [www.cs.ucsb.edu]
- [www.cs.ucsb.edu]
- [www.ic.unicamp.br]
- [cs.ucsb.edu]
- [research.microsoft.com]
- [research.microsoft.com]
- [research.microsoft.com]
- [research.microsoft.com]
- [research.microsoft.com]
- [research.microsoft.com]
- [cs.ucsb.edu]
- DBLP
Other Repositories/Bibliography
| Venue: | In Infocom’03 |
| Citations: | 166 - 14 self |
BibTeX
@INPROCEEDINGS{Castro03anevaluation,
author = {Miguel Castro and Michael B. Jones and Anne-Marie Kermarrec and Antony Rowstron and Marvin Theimer and Helen Wang and Alec Wolman},
title = {An Evaluation of Scalable Application-Level Multicast Built Using Peer-to-Peer Overlays},
booktitle = {In Infocom’03},
year = {2003}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Structured peer-erg163 overlay networks such as CAN, Chord, Pastry, and Tapestry can be used to implement Internet-g683 application-3 vel multicast. There are two general approaches to accomplishingthis: tree buildingand flooding. This paper evaluates these two approaches usingtwo different types of structured overlay: 1) overlays which use a form of generalized hypercube routing, e.g., Chord, Pastry and Tapestry, and 2) overlays which use a numerical distance metric to route through a Cartesian hyper-erg15 e.g., CAN. Pastry and CAN are chosen as the representatives of each type of overlay. To the best of our knowledge, this paper reports the firstheadto -d- comparison ofCAN-B91g versus Pastry-gZ4B overlay networks, usingmulticast communication workloads runningon an identical simulation infrastructure. The two approaches to multicast are independent of overlay network choice, and we provide a comparison of floodingversus tree-2696 multicast on both overlays. Results show that the tree-2613 approach consistently outperforms the floodingapproach. Finally, for treebased multicast, we show that Pastry provides better performance than CAN.
Keyphrases
identical simulation infrastructure numerical distance structured overlay internet-g683 application-3 vel multicast tree buildingand flooding approach usingtwo different type cartesian hyper-erg15 overlay network choice generalized hypercube routing peer-erg163 overlay network floodingversus tree-2696 multicast tree-2613 approach general approach usingmulticast communication workload
