Áika: A Distributed Edge System for AI Inference

Video monitoring and surveillance of commercial fisheries in world oceans has been proposed by the governing bodies several nations as a response to crimes such overfishing. Traditional video systems may not be suitable due limitations offshore fishing environment, including low bandwidth, unstable satellite network connections issues preserving privacy crew members. In this paper, we present Áika, robust system for executing distributed Artificial Intelligence (AI) applications on edge. Áika provides engineers researchers with building blocks form Agents, which enable expression computation pipelines robustness guarantees. Agents are continuously monitored dedicated nodes, provide checkpointing replication scheme. is designed privacy-sensitive environments, where flexible access policies at storage level can guarantees data transfer access.