Enabling Scalable Real-Time Sensor Stream Processing Across Decentralized Privacy-Preserving Storage Solutions

Internet of Things (IoT), Smart Health, and Smart City applications generate continuous data streams that may contain sensitive personal information. Decentralized storage systems such as Solid provide data ownership, access control, and interoperable sharing but offer limited support for real-time stream processing. We present Heimdall, an intermediary analytics service that registers RSP-QL queries over streams stored in Solid Pods to produce query results and reuses existing query executions when supported reuse conditions are satisfied. We evaluate Heimdall against Client-Side Processing and a notification intermediary using a wearable-sensor workload with concurrent client scaling, query and data heterogeneity, and concurrent non-reusable queries. For equivalent queries, Heimdall maintains nearly constant latency as client count increases, while Client-Side Processing shows substantial latency growth and instability at higher concurrency. Heimdall also reduces accumulated CPU and memory consumption and client-side network traffic by sharing stream retrieval and query execution. When query execution cannot be reused, shared stream acquisition still improves scalability, although degradation appears as the number of independent queries and physical streams increases with no-reuse. These results show that shared stream acquisition and continuous-query execution can reduce duplicated computation and communication in decentralized stream processing.