A Dynamic, Decentralised Search Algorithm for Efficient Data Retrieval in a Distributed Tuple Space

Atkinson, A.

This paper presents an algorithm which may be used to efficiently search for and retrieve tuples in a distributed tuple space. The algorithm, a core part of the Tuple- ware system, is based on the success or failure of previous tuple requests to remote nodes in the system, and this data is used determine the relative probability of particular remote nodes being able to fulfil subsequent future requests. The logic of this algorithm is distributed and decentralised: each node dynamically calculates its relationship with other nodes at runtime. The behaviour of the algorithm using two applications is analysed, and shows significant improvement in terms of efficiency and performance compared to comparable tuple space implementations.

Cite as: Atkinson, A. (2010). A Dynamic, Decentralised Search Algorithm for Efficient Data Retrieval in a Distributed Tuple Space. In Proc. Eighth Australasian Symposium on Parallel and Distributed Computing (AusPDC 2010) Brisbane, Australia. CRPIT, 107. Chen, J. and Ranjan, R. Eds., ACS. 21-29

(from crpit.com) (local if available)