Search

Data fragmentation has recently emerged as a complementary approach to encryption for protecting confidentiality of sensitive associations when storing data at external parties. In this paper, we discuss how the use of indexes, typically associated with the encrypted portion of the data, while desirable for providing effectiveness and efficiency in query execution, can - combined with fragmentation - cause potential leakage of confidential (encrypted or fragmented) information. We illustrate how the exposure to leakage varies depending on the kind of indexes. Such observations can result useful for the design of approaches assessing information exposure and for the definition of safe (free from inferences) indexes in fragmented data.

Traditional access control solutions, based on preliminary identification and authentication of the access requester, are not adequate for the context of open web service systems, where servers generally do not have prior knowledge of the requesters. The research community has acknowledged such a paradigm shift and several investigations have been carried out for new approaches to regulate access control in open dynamic settings. Typically based on logic, such approaches, while appealing for their expressiveness, result not applicable in practice, where simplicity, efficiency, and consistency with consolidated technology are crucial. The eXtensible Access Control Markup Language (XACML) has established itself as the emerging technological solution for controlling access in an interoperable and flexible way. Although supporting the most common policy representation mechanisms and having acquired a significant spread in the research community and the industry, XACML still suffers from some limitations which impact its ability to support actual requirements of open web-based systems. In this paper, we provide a simple and effective formalization of novel concepts that have to be supported for enforcing the new access control paradigm needed in open scenarios, toward the aim of providing an expressive solution actually deployable with today's technology. We illustrate how the concepts of our model can be deployed in the XACML standard by exploiting its extension points for the definition of new functions, and introducing a dialog management framework to enable access control interactions between web service clients and servers.

Nowadays, users are more and more exploiting external storage and connectivity for sharing and disseminating user-generated content. To this aim, they can benefit of the services offered by Internet companies, which however assume that the service provider is entitled to access the resources. To overcome this limitation, we present an approach that does not require complete trust in the external service w.r.t. both resource content and authorization management, while at the same time allowing users to delegate to the provider the enforcement of the access control policy on their resources. Our solution relies on the translation of the access control policy into an equivalent encryption policy on resources and on a hierarchical key structure that limits both the number of keys to be maintained and the amount of encryption to be enforced.

The balance between privacy and utility is a classical problem with an increasing impact on the design of modern information systems. On the one side it is crucial to ensure that sensitive information is properly protected; on the other side, the impact of protection on the workload must be limited as query efficiency and system performance remain a primary requirement. We address this privacy/efficiency balance proposing an approach that, starting from a flexible definition of confidentiality constraints on a relational schema, applies encryption on information in a parsimonious way and mostly relies on fragmentation to protect sensitive associations among attributes. Fragmentation is guided by workload considerations so to minimize the cost of executing queries over fragments. We discuss the minimization problem when fragmenting data and provide a heuristic approach to its solution.

Several architectures have been recently proposed that store relational data in encrypted form on untrusted relational databases. Such architectures permit the creation of novel Internet services and also offer an opportunity for a better construction of ASP solutions. Environments where there are limited resources that do not permit an efficient management of databases or where it is critical to offer a robust Internet access to private data may all benefit from the above architectures. In this paper we analyze the impact that this architecture has on the typical services of a database. The analysis is based on the experience gained in the construction of a prototype of a complete architecture for the management of encrypted databases. Specifically, we illustrate the impact on query translation and optimization, and the main components of the software architecture of the prototype.

Current P2P (overlay) networks have been designed with a focus on the construction of efficient mechanisms for the storage and retrieval of information among a community of participating nodes, with limited attention to the protection from malicious behaviours. The paper analyzes the attacks that can be realized on the choice of identifiers by the nodes and proposes adequate countermeasures. The paper uses as a reference the Kademlia protocol, but most of the results are immediately applicable to every DHT network.

Presents a technique for analyzing the run-time behavior of integrity constraint repair actions, i.e. active database rules that are specifically designed to correct violations of database integrity. When constraints become violated due to an incorrect user transaction, rule computation is started to restore the database to a correct state. Since repair actions may be numerous and may conflict with each other, automated support for the analysis of their run-time behavior is necessary. The proposed technique helps the rule base administrator define a repair rule selection strategy such that the computation terminates for every input transaction, the final database state satisfies all the constraints, and the user's preferences among different ways to restore integrity are taken into account. In addition, it can be used by the rule designer to spot "dangerous" rules that may be subject to redesign. This problem is formulated as an optimization problem on directed hypergraphs, which we demonstrate to be NP-hard and which we solve by means of a heuristic algorithm.