Efficient filtering for massively distributed video games

Luciana Arantes 1, 2, * Maria Potop-Butucaru 1, 2, * Pierre Sens 1, 2, * Mathieu Valero 1, 2, *
* Corresponding author
2 Regal - Large-Scale Distributed Systems and Applications
LIP6 - Laboratoire d'Informatique de Paris 6, Inria Paris-Rocquencourt
Abstract : Distributed R-trees overlays recently emerged as an alternative for efficiently implementing DHT-free publish/subscribe communication primitives. Overlays using R-tree index structures offer logarithmic delivery garanties, guarantee zero false negatives and considerably reduce the number of false positives. In this paper we extend the distributed R-trees (DR-Trees) in order to meet two key requirements in massively distributed video game applications: load balancing and low latency. Our optimizations target both thestructural organisation of the DR-Trees and the publication policies. The contribution of the current work steams in an extensive evaluation of the novel structure along four parameters: latency, load, scalability and the rate of false positives. Interestingly, the novel structure performs better than the traditional distributed R-tree both in terms of load balancing and latency. Additionally, it does not alter the performances related to the scalability and the rate of false positives and negatives a node has to filter.
Complete list of metadatas

https://hal.inria.fr/inria-00408209
Contributor : Pierre Sens <>
Submitted on : Wednesday, July 29, 2009 - 4:00:30 PM
Last modification on : Thursday, March 21, 2019 - 2:18:09 PM
Long-term archiving on: Tuesday, June 15, 2010 - 6:48:42 PM

File

RR-7008.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00408209, version 1

Citation

Luciana Arantes, Maria Potop-Butucaru, Pierre Sens, Mathieu Valero. Efficient filtering for massively distributed video games. [Research Report] RR-7008, INRIA. 2009, pp.19. ⟨inria-00408209⟩

Share

Metrics

Record views

504

Files downloads

134