Beekup: A distributed and safe P2P storage framework for IoE applications

• Main Authors: Rizzo, Federico, Spoto, Giovanni Luca, Brizzi, Paolo, Bonino, Dario, Di Bella, Giuseppe, Castrogiovanni, Pino
• Format: Conference Proceeding
• Language: English
• Subjects: Beekup; Cloud computing; Cloud storage; Community cloud; Distributed file system; Edge storage; IoE storage; Memory; Peer-to-peer computing; Reliability; Scalability; Secure storage; Sensors
• ISSN: 2472-8144
• DOI: 10.1109/ICIN.2017.7899248

Cloud computing and, in particular, cloud storage, has successfully boosted a variety of applications, in several domains; with proven acceptance and benefits by end users. While peculiar features of the cloud, e.g., redundancy, scalability and accessibility, are still highly desirable, several new scenarios are nowadays emerging, which call for additional features and/or different interpretations of the underlying paradigm. With the recent rising of the Internet of Everything, for example, some of the cloud principles should be partly reconsidered to better address scenarios involving a high number of devices, each carrying limited storage capacity, which can possibly operate in connection "islands" due to erratic or unreliable network connections. In this paper we start addressing these needs by presenting the Beekup system: a P2P storage framework, based on Tahoe-LAFS. Beekup enables cloud-like storage on edge devices supporting a variety of applications: from private secure file backup, to sensor-level distributed storage of raw data. Two main testbeds centered on private cloud storage and distributed sensor-data backup are analyzed in the paper, together with corresponding user applications. Initial outcomes from experiments and encouraging feedback from users involved in system trials, confirm the potential viability of the approach.