Adaptive Stream Processing on Edge Devices through Active Inference

TL;DR

This paper introduces a novel Active Inference-based machine learning approach for adaptive stream processing on edge devices, improving SLO fulfillment, transparency, and convergence speed in IoT scenarios.

Contribution

It presents the first application of Active Inference to edge stream processing, enabling autonomous, transparent, and efficient management of multiple SLOs in heterogeneous environments.

Findings

Converges to optimal solutions within 30 iterations

Guarantees transparency in decision-making

Achieves accurate SLO fulfillment in real-time

Abstract

The current scenario of IoT is witnessing a constant increase on the volume of data, which is generated in constant stream, calling for novel architectural and logical solutions for processing it. Moving the data handling towards the edge of the computing spectrum guarantees better distribution of load and, in principle, lower latency and better privacy. However, managing such a structure is complex, especially when requirements, also referred to Service Level Objectives (SLOs), specified by applications' owners and infrastructure managers need to be ensured. Despite the rich number of proposals of Machine Learning (ML) based management solutions, researchers and practitioners yet struggle to guarantee long-term prediction and control, and accurate troubleshooting. Therefore, we present a novel ML paradigm based on Active Inference (AIF) -- a concept from neuroscience that describes how the brain constantly predicts and evaluates sensory information to decrease long-term surprise. We implement it and evaluate it in a heterogeneous real stream processing use case, where an AIF-based agent continuously optimizes the fulfillment of three SLOs for three autonomous driving services running on multiple devices. The agent used causal knowledge to gradually develop an understanding of how its actions are related to requirements fulfillment, and which configurations to favor. Through this approach, our agent requires up to thirty iterations to converge to the optimal solution, showing the capability of offering accurate results in a short amount of time. Furthermore, thanks to AIF and its causal structures, our method guarantees full transparency on the decision making, making the interpretation of the results and the troubleshooting effortless.