SOUP is a novel decentralized online social network (DOSN).
It allows its users to regain control over their own data and share what they want with whom they want.
At its core is a new, generic approach to placing replicas of
user data in a decentralized OSN, that can also be applied to other OSNs besides SOUP.
In addition to enabling every user
to store their data at their own machine, SOUP—with the
help of the user’s friends’ recommendations—efficiently
distributes replicas of the user’s data to cleverly selected
OSN participants. SOUP does not rely on permanently
available or altruistically provided storage, although it is
able to make an opportunistic use of such resources as they
become available. Despite conservative assumptions on the
availability of resources, SOUP’s performance in terms of
data availability is close to that of a centralized solution. By
introducing little replica overhead and low communication
overhead, SOUP also scales well with large OSNs.
It further provides salient key features, which distinguish our approach from related works.
It is robust as a whole: regardless of participants’ social relations or online
probabilities, data of all participants will be highly available.
It is adaptive to the dynamics often seen in a decentralized
OSN—such as frequent node joining and departure— and it
can quickly respond to changes in the system and continue
to provide high performance. SOUP is also resilient against
a high ratio of malicious OSN users without a significant
performance degradation; even if half of the identities in an
OSN are controlled by an adversary and attack SOUP, the
system will perform steadily.
To ensure the privacy of user data, we encrypt all private data of each user with Attribute Based Encryption (ABE).
This provides the users with fine-grained access control to their data.
For detailed information on the architecture, system design and our replica placing algorithm please refer to the publications section.