A cross-platform data mart synchronization model for high availability in dual-cloud architectures
Babawale Patrick Okare · Journal of Advanced Education and Sciences · 2021
In the evolving landscape of enterprise data management, dual-cloud architectures have become a strategic choice to enhance redundancy, compliance, and vendor flexibility. However, ensuring high availability and consistent data synchronization across heterogeneous cloud platforms poses significant technical challenges, particularly for distributed data marts supporting critical analytical workloads. This paper presents a comprehensive, platform-agnostic synchronization model designed to address these challenges by harmonizing disparate storage engines, replication protocols, and metadata schemes. The proposed layered architecture incorporates real-time Change Data Capture mechanisms, bidirectional synchronization logic, and advanced conflict resolution strategies to maintain near-real-time data consistency. Embedded monitoring, automated recovery workflows, and governance controls further strengthen system reliability and compliance. The model facilitates seamless failover and resilience against cloud-specific outages and regional failures, while optimizing operational efficiency through incremental sync and smart batching. By bridging theoretical foundations with practical implementation considerations, this work advances the state of cross-cloud data mart synchronization and supports modern multi-cloud data strategies. Future research directions include AI-enhanced conflict resolution and empirical validation of synchronization performance, promising to extend the model’s applicability and robustness in dynamic data environments.