Achlys: Towards a Framework for Distributed Storage and Generic Computing Applications for Wireless IoT Edge Networks with Lasp on GRiSP

摘要: Internet of Things (IoT) continues to grow exponentially, in number devices and the amount data they generate. Processing this requires an exponential increase computing power. For example, aggregation can be done directly at edge. However, is very limited; ideally we would like do more general computations In paper propose a framework for doing general-purpose edge on sensor networks themselves, without requiring external connections gateways or cloud. This challenging because have unreliable communication, nodes, limited (if any) power storage. How implement production-quality components these networks? We need bridge gap between unreliable, infrastructure stringent requirements components. To solve problem present Achlys, that provides reliable storage, computation, communication capabilities wireless IoT nodes. Using network able configure manage itself directly, connectivity. Achlys combines Lasp key/value store Partisan library. efficient decentralized storage based properties CRDTs (Conftict-Free Replicated Data Types). connectivity broadcast hybrid gossip. Both are specifically designed extremely resilient. They continue working despite high node churn, frequent partitions, communication. Our first implementation GRiSP embedded system boards. choose as our platform it implements high-level functionality, namely Erlang, bare hardware supports Pmod sensors give some results using building systems explain how plan evolve future. work progress being context LightKone European H2020 research project, process implementing evaluating proof-of-concept application area precision agriculture.