WedgeChain: A Trusted Edge-Cloud Store With Asynchronous (Lazy) Trust

Abstract: We propose WedgeChain, a data store that spans both edge and cloud nodes (an edge-cloud system). WedgeChain consists of a logging layer and a data indexing layer. In this study, we encounter two challenges: (1) edge nodes are untrusted and potentially malicious, and (2) edge-cloud coordination is expensive. WedgeChain tackles these challenges by the following proposals: (1) Lazy (asynchronous) certification: where data is committed at the untrusted edge and then lazily certified at the cloud node. This lazy certification method takes advantage of the observation that an untrusted edge node is unlikely to act maliciously if it knows it will be detected (and punished) eventually. Our lazy certification method guarantees that malicious acts (i.e., lying) are eventually detected. (2) Data-free certification: our lazy certification method only needs to send digests of data to the cloud, instead of sending all data to the cloud, which enables saving network and cloud resources and reduce costs. (3) LSMerkle: we extend a trusted index (mLSM) to enable indexing data at the edge while utilizing lazy and data-free certification.