Reducing transaction abort rate of epidemic algorithm in replicated databases

Easy to deploy, robust, and highly resilient to failures, epidemic algorithms are a potentially effective mechanism for propagating information in large peer-to-peer systems deployed on Internet or ad hoc networks. In the paper, we explore the epidemic algorithms used for transaction processing in replicated databases that reside in weakly connected environments. We concentrate on the transaction commit voting process of the epidemic algorithms and suggest a new voting method, which takes an optimistic approach in conflict reconciliation. The optimistic voting protocol decreases abort rate and improves average response time of transactions.