Designing Reconfigurable Intelligent Systems with Markov Blankets
This addresses the challenge of efficient and scalable system management in distributed computing environments, though it appears incremental as it builds on existing decentralization concepts with a specific causality-based approach.
The paper tackled the problem of managing large-scale Compute Continuum systems by decentralizing the evaluation of Service Level Objectives (SLOs) and device reconfiguration, using a causality filter based on Markov blankets to reduce communication overhead and enable devices to perceive and act on their environment for Quality of Service (QoS) assurance.
Compute Continuum (CC) systems comprise a vast number of devices distributed over computational tiers. Evaluating business requirements, i.e., Service Level Objectives (SLOs), requires collecting data from all those devices; if SLOs are violated, devices must be reconfigured to ensure correct operation. If done centrally, this dramatically increases the number of devices and variables that must be considered, while creating an enormous communication overhead. To address this, we (1) introduce a causality filter based on Markov blankets (MB) that limits the number of variables that each device must track, (2) evaluate SLOs decentralized on a device basis, and (3) infer optimal device configuration for fulfilling SLOs. We evaluated our methodology by analyzing video stream transformations and providing device configurations that ensure the Quality of Service (QoS). The devices thus perceived their environment and acted accordingly -- a form of decentralized intelligence.