This research addresses the challenge of managing latency in distributed computer systems. Maintaining correct delays in data transmission across various network conditions is crucial for system efficiency and security. We focus on improving the Adaptive Wide-Area Replication (AWARE) algorithm, a method used to coordinate data across different locations in a way that minimizes delays. To enhance AWARE, we incorporate two concepts: Vivaldi network coordinates and New- tonian invariants. Vivaldi network coordinates help the system better understand and calculate the physical layout of the network by embedding network members in Euclidean space, where the distance in this space represents latency be- tween them. Newton invariants are rules based on physics that help the system detect and adjust for any unusual changes in network latency that might be caused by technical issues or security threats. We evaluated the original and enhanced AWARE algorithms by simulating typical network operations and various attack scenarios designed to slow down the process of reaching consensus. Our findings show that the enhanced AWARE algorithm provides more accurate and robust management of network latency, especially under attack conditions, leading to a more reliable and secure distributed system. This study confirms that integrating correction techniques into latency management processes significantly improves the resilience and accuracy of distributed systems.