Arianna Dragoni

This paper proposes Ermes, a distributed platform that natively integrates state management into the FaaS paradigm, enabling the joint distribution of computational workloads and application state across the edge-to-cloud continuum.

This work introduces a Binary Linear Programming model to compute optimal placements and proposes a topology-aware greedy heuristic that efficiently approximates the optimal solution, making it suitable for periodic system reconfigurations.

WASP enables seamless adaptation of serverless applications to diverse edge-cloud environments without the need for code changes, unlocking new possibilities for stateful workloads.

Data placement, not function scheduling, is the key determinant of performance in edge-cloud serverless architectures, with a heuristic that scales to over 10,000 nodes while keeping latency stable.