Scalable DDS Discovery Protocols Based on Bloom Filters

Research areas:

• Methodology - Machine learning

Year:

2007

Type of Publication:

In Proceedings

Authors:

• Sánchez-Monedero, Javier
• Lopez-Soler, J. M.
• Povedano-Molina, J.

Book title:

Real-time and Embedded Systems Workshop, Arlington VA, USA

Abstract:

The OMG Data Distribution Service (DDS) standard has been recently adopted for real-time data communications. In any DDS platform, to deal with heterogeneous environments and to provide transparent communications, an entity discovery protocol is required. Broadly speaking, an exhaustive discovery approach could result in quadratic complexity. Obviously, the scalability of DDS infrastructure is somehow limited by the adopted discovery protocol. In this work, inspired by peer-to-peer and agent based networking services, we propose a preliminary study on scalable discovery protocols for DDS. In particular, we propose and evaluate solutions to the discovery DDS scalability problem by using Bloom Filters (BF).