Content Evacuation in Inter-DC Optical Networks under Post-Disaster Cascading Failures

In the post-pandemic era, global work patterns have been reshaped, and the demand for cloud migration for enterprises and government has increased. As a result, cloud data disaster backup/recovery technology has been gaining more attention. Moving beyond the traditional focus on pre-disaster content backup, our study addresses the challenge of rapidly evacuating content during cascading failures in post-disaster scenarios. Due to the interdependence of i) data centers (DCs), ii) inter-DC optical networks, and iii) power grid networks, disasters can have a domino effect on these infrastructures, with their impact gradually expanding over time and space. In this work, we propose two trajectory models for constructing the spatio-temporal features of the inter-DC optical network under cascading failures, and we propose a trajectory-based content evacuation strategy (TCE). Numerical results show that TCE can reduce content loss by up to 25% compared to baseline content evacuation strategies.