2015/02/16 by Marcus Brandenburger, Christian Cachin, Brandenburger, Marcus +3
Computer Science · #Advanced Data Storage Technologies #Cloud Computing and Resource Management #Cryptography and Security (cs.CR) #Distributed #Distributed systems and fault tolerance #FOS: Computer and information sciences #Parallel #and Cluster Computing (cs.DC)
paper · pdf · doi:10.48550/arxiv.1502.04496
openalex publication_date 2015/02/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Cloud services have turned remote computation into a commodity and enable convenient online collaboration. However, they require that clients fully trust the service provider in terms of confidentiality, integrity, and availability. Towards reducing this dependency, this paper introduces a protocol for verification of integrity and consistency for cloud object storage (VICOS), which enables a group of mutually trusting clients to detect data-integrity and consistency violations for a cloud object-storage service. It aims at services where multiple clients cooperate on data stored remotely on a potentially misbehaving service. VICOS enforces the consistency notion of fork-linearizability, supports wait-free client semantics for most operations, and reduces the computation and communication overhead compared to previous protocols. VICOS is based in a generic way on any authenticated data structure. Moreover, its operations cover the hierarchical name space of a cloud object store, supporting a real-world interface and not only a simplistic abstraction. A prototype of VICOS that works with the key-value store interface of commodity cloud storage services has been implemented, and an evaluation demonstrates its advantage compared to existing systems.