Databases Reference
In-Depth Information
6
Towards Secure Data Outsourcing
Radu Sion
Network Security and Applied Cryptography Lab
Computer Science, Stony Brook University
sion@cs.stonybrook.edu
Summary.
The networked and increasingly ubiquitous nature of today's data man-
agement services mandates assurances to detect and deter malicious or faulty be-
havior. This is particularly relevant for outsourced data frameworks in which clients
place data management with specialized service providers. Clients are reluctant to
place sensitive data under the control of a foreign party without assurances of confi-
dentiality. Additionally, once outsourced, privacy and data access correctness (data
integrity and query completeness) become paramount. Today's solutions are fun-
damentally insecure and vulnerable to illicit behavior, because they do not handle
these dimensions.
In this chapter we will explore the state of the art in data outsourcing mech-
anisms providing strong security assurances of (1)
correctness
,(2)
confidentiality
,
and (3) data access
privacy
.
There exists a strong relationship between such assurances; for example, the lack
of access pattern privacy usually allows for statistical attacks compromising data
confidentiality. Confidentiality can be achieved by data encryption. However, to be
practical, outsourced data services should allow expressive client queries (e.g., rela-
tional joins with arbitrary predicates) without compromising confidentiality. This is
a hard problem because decryption keys cannot be directly provided to potentially
untrusted servers. Moreover, if the remote server cannot be fully trusted, protocol
correctness become essential.
Here we will discuss query mechanisms targeting outsourced relational data that
(i) ensure queries have been executed with integrity and completeness over their
respective target data sets, (ii) allow queries to be executed with confidentiality
over encrypted data, (iii) guarantee the privacy of client queries and data access
patterns. We will then propose protocols that adapt to the existence of
trusted
hardware
— so critical functionality can be delegated securely from clients to servers.
We have successfully started exploring the feasibility of such solutions for providing
assurances for query execution and the handling of binary predicate JOINs with full
privacy in outsourced scenarios.
The total cost of ownership of data management infrastructure is 5-10 times
greater than the hardware costs, and more data is produced and lives digitally every
day. In the coming years, secure, robust, and ecient outsourced data management
will be demanded by users. It is thus important to finally achieve outsourced data
