APOLLO: A Proximity-Oriented Low-Level Orchestration Algorithm for Performance Optimization in Mist Computing

Abstract The fusion of satellite technologies with the Internet of Things (IoT) has propelled the evolution of mobile computing, ushering in novel communication paradigms and data management strategies. Within this landscape, the efficient management of computationally intensive tasks in satellite-enabled mist computing environments emerges as a critical challenge. These tasks, spanning from optimizing satellite communication to facilitating blockchain-based IoT processes, necessitate substantial computational resources and timely execution. To address this challenge, we introduce APOLLO, a novel low-layer orchestration algorithm explicitly tailored for satellite mist computing environments. APOLLO leverages proximity-driven decision-making and load balancing to optimize task deployment and performance. We assess APOLLO's efficacy across various configurations of mist layer devices while employing a round-robin principle for equitable task distribution among the close low-layer satellites. Our findings underscore APOLLO's promising outcomes in terms of reduced energy consumption, minimized end-to-end delay, and optimized network resource utilization, particularly in targeted scenarios. However, the evaluation also reveals avenues for refinement, notably in CPU utilization and slightly low tasks success rates. Our work contributes substantial insights into advancing task orchestration in satellite-enabled mist computing with more focus on energy and end-to-end sensitive applications, paving the way for more efficient, reliable, and sustainable satellite communication systems.