Fernández de Barrena Sarasola, TelmoGarcía Gangoiti, AnderFerrando, Juan Luis2025-02-262025-02-262024De Barrena Sarasola, T. F., García, A., & Ferrando, J. L. (2024). IIoT Protocols for Edge/Fog and Cloud Computing in Industrial AI: A High Frequency Perspective. International Journal of Cloud Applications and Computing, 14(1), 1-30. https://doi.org/10.4018/IJCAC.3421282156-183410.4018/IJCAC.342128https://hdl.handle.net/20.500.14454/2381Various industrial applications deal with high-frequency data. Traditionally, these systems have analyzed high-frequency data directly on the data source or at the commanding PLC. However, currently, Industry 4.0 technologies support new monitoring scenarios for high-frequency data monitoring where raw data is transmitted in soft-real time to an Edge/Fog or Cloud node for processing, enabling centralized computing. This demands efficient communication protocols capable of handling high-frequency, high-throughput data. This paper focuses on analyzing the performance of key IIoT (Industrial Internet of Things) messaging protocols-AMQP, MQTT, KAFKA, ZeroMQ, and OPCUA-to evaluate their suitability, in terms of latency and jitter, for transmitting high-frequency data within these new scenarios. The analysis reveals MQTT, AMQP, and ZeroMQ as top performers in Edge/Fog computing, while ZeroMQ exhibits the lowest latency and jitter in Cloud computing. Finally, a guideline for protocol selection is proposed, aiding industrial enterprises in protocol selection for specific AI use cases.engCloud computingEdge/Fog computingHigh frequency signalsIIoT protocolsIndustry 4.0journal article2025-02-262156-1826