In this thesis, I propose an algorithm for optimizing the cost involved with synchronizing the data contained in mobile wireless devices, named Expedient Trickle Sync (ETS). In this thesis, I focus on two significant cost factors: firstly, that of the cost of transmitting information across the network, and secondly, the cost associated with user access to out-of-date information. The ETS algorithm attempts to balance these two cost factors via a simple set of heuristics which calculate at regular intervals a test value, based on a variety of observations, and a threshold value which is recalculated on a daily basis. Additionally, the ETS algorithm will prioritize records based on their probability of access and the cost associated with synchronizing them, thus possibly deferring the synchronization of records with a low probability of access until it is possible to resolve their replication in a lower-cost network environment. This thesis evaluates the ETS algorithm alongside other optimistic replication algorithms in a custom developed simulation environment, the results of which show that in many use scenarios, the ETS algorithm can indeed reduce the cost of data synchronization.
| Identifer | oai:union.ndltd.org:uvic.ca/oai:dspace.library.uvic.ca:1828/1335 |
| Date | 17 January 2009 |
| Creators | Barclay, Brad |
| Contributors | Weber, Jens |
| Source Sets | University of Victoria |
| Language | English, English |
| Detected Language | English |
| Type | Thesis |
| Rights | Available to the World Wide Web |
Page generated in 0.0022 seconds