Abstract Data replication is used in large distributed systems to provide high fault tolerance and improve performance. However, when replication is used, the data becomes vulnerable to inconsistency issues. Replication control protocols (RCPs) are needed to “synchronize” concurrent read (queries) and write (writes) operations on the replicated data. The purpose of this document is to ask if a single protocol can... Data replication is used in large distributed systems to provide high fault tolerance and improve performance. However, when replication is used, the data becomes vulnerable to inconsistency issues. Replication control protocols (RCPs) are needed to “synchronize” concurrent read (queries) and write (writes) operations on the replicated data. The purpose of this document is to ask if a single protocol can be provided that can be implemented in any logical two-dimensional structure and to provide optimal performance concerning communication cost, availability, and system load for its read/write operations. The proposed structured replication control protocol (A2DS) can be implemented with any logical 2-dimensional structure. Read More Read Less

Abstract This work marks a significant step in the modernization of interconnected systems. We introduce cost-efficient, standardized, and modular components, a novel approach in this context. For the first time, we implement Commercial Off-The-Shelf (COTS) products in an interconnected system using a distributed architecture. This innovative use of COTS products is expected to lead to further cost savings and significant efficiency... This work marks a significant step in the modernization of interconnected systems. We introduce cost-efficient, standardized, and modular components, a novel approach in this context. For the first time, we implement Commercial Off-The-Shelf (COTS) products in an interconnected system using a distributed architecture. This innovative use of COTS products is expected to lead to further cost savings and significant efficiency improvements. Read More Read Less

Abstract Carpooling can reduce traffic congestion and the environmental impact of car use. This paper addresses an important problem for dynamic carpooling: calculating the shortest drivers’ and passengers’ paths. These paths are synchronized in that they have a common sub-path between 2 points: the point where the driver picks up the passenger and the point where the passenger drops off the... Carpooling can reduce traffic congestion and the environmental impact of car use. This paper addresses an important problem for dynamic carpooling: calculating the shortest drivers’ and passengers’ paths. These paths are synchronized in that they have a common sub-path between 2 points: the point where the driver picks up the passenger and the point where the passenger drops off the car. A passenger path may include time-dependent public transportation elements before or after this common sub-path. This defines the 2SynchronizationPointShortestPath Problem (2SSPPP). We demonstrate that 2SPSPP is a polynomial problem with a worst-case solution. However, efficient algorithms must solve this problem in realistic transportation networks. This paper focuses on efficiently calculating optimal itineraries to solve this 2SPSPP problem, i.e., determining (optimal) pick-up/drop-off points and two synchronized paths that minimize total travelling time. Restriction areas are defined for reasonable pickup/drop-off points and are used to guide the algorithms using landmarks-based heuristics. Experiments are performed on real transportation networks. The results demonstrate the performance of the suggested algorithms and the CPU time and application interest of the restriction areas for pick-up and drop-off points. Read More Read Less