The increasing complexity of modern electrical power systems requires advanced monitoring infrastructures capable of guaranteeing low latency, scalability, and resilience across geographically distributed environments. In this context, Phasor Data Concentrators (PDCs) play a crucial role in aggregating and forwarding time-synchronized measurements collected by Phasor Measurement Units (PMUs). The placement of PDCs within the communication network directly affects end-to-end latency, bandwidth utilization, and overall system stability. This thesis addresses the problem of latency-aware PDC placement through a twofold approach that combines graph-based optimization models with cloud-native multi-cluster orchestration in a computing continuum perspective. First, the monitoring network is modeled as a constrained graph, where nodes and links are characterized by latency, bandwidth, and availability attributes.