You are looking at a specific version 20191021:082011 of this paper. See the latest version.

Paper 2019/1223

Integrita: Protecting View-Consistency in Online Social Network with Federated Servers

Sanaz Taheri Boshrooyeh and Alptekin Küpçü and Öznur Özkasap

Abstract

In the current designs of Online Social Networks (OSN), like Facebook and Twitter, with a central service provider, users' read and write requests over the shared data (e.g., Facebook wall or a group page) are handled via a central OSN provider. However, such centralization comes with view consistency issues where a corrupted provider may serve users with different views of the shared data e.g., by adding, dropping or reordering posts. Integrita provides a data-sharing platform that empowers view consistency relying on N federated servers whose N-1 can be malicious and colluding. Users are guaranteed that the servers cannot show divergence view of the shared data (e.g., posts of the group page) to the users (e.g., group members) without being detected. Unlike the state-of-the-art, Integrita enables detection of inconsistency neither by using storage inefficient data replication solution nor by requiring users to exchange their views out of the band. Every user, without relying on the presence of other users, can verify any server-side equivocation regarding her performed operation. We introduce and achieve a new level of view consistency called q-detectable consistency in which any inconsistency between users' view cannot remain undetected for more than q posts. The data-sharing platform of Integrita advances the centralized and distributed counterparts by improving the view-consistency and storage overhead (by the factor of 1/N where N is the number of the servers), respectively. Nevertheless, concerning per server storage overhead and cross-server communication, Integrita's overhead is the minimum among all its counterparts.

Metadata
Available format(s)
PDF
Category
Applications
Publication info
Preprint. MINOR revision.
Keywords
View consistencyq-Detectable ConsistencyStrong ConsistencyMaliciousShared DataCollaborative Data SharingIntegrityHistory Integrity.
Contact author(s)
staheri14 @ ku edu tr
History
2021-02-27: last of 3 revisions
2019-10-21: received
See all versions
Short URL
https://ia.cr/2019/1223
License
Creative Commons Attribution
CC BY
Note: In order to protect the privacy of readers, eprint.iacr.org does not use cookies or embedded third party content.