Smart systems are characterised by their ability to analyse measured data in live and to react to changes according to expert rules. Therefore, such systems exploit appropriate data models together with actions triggered by domain-related conditions. Often thousands of updates need to be processed in a short amount of time, to detect which rules need to be triggered. The challenge at hand is that smart systems usually run on restricted hardware, like Raspberry Pi. There have been several approaches investigated to efficiently check conditions on data models, such as OCL-Gremlin or EMF-IncQuery. However, these solutions assume either that rules and data models fit into main memory or rely on high latency persistence storages, which severely damage the reactivity of smart systems. To tackle this challenge, we propose a novel composition process, which weaves executable rules into a data model with lazy loading abilities. Our approach targets models@run.time usages and has been evaluated on a real-world home automation case study. We quantitatively show that our approach can handle, at low latency, big sets of rules on top of large-scale data models, even on restricted hardware.