Consistency-Aware Multi-Server Network Design for Delay-Sensitive Applications Under Multiple-Server Failures

Delay-sensitive applications demand minimal latency and strict event ordering. Distributing processing to multiple edge servers can reduce access delay, but it is essential to ensure a consistent state across those servers and handle server failures. Existing models address consistency with only single-server failures, but multiple-server failures can occur in real networks. This paper proposes a server allocation model for a consistency-aware multi-server network in delay-sensitive applications with preventive start-time optimization (PSO) under multiple-server failures. PSO determines the assignment to minimize the worst-case delay across all considered failure scenarios without introducing service disruption for users connected to non-failed servers. We formulate the proposed model as an integer linear programming problem. We prove the decision version of the server allocation problem is NP-hard. Numerical results show that the proposed model outperforms start-time optimization in terms of the largest total delay and run-time optimization in terms of avoiding instability.