Graduate Thesis Or Dissertation

 

Efficient Oblivious Access to Trees Público Deposited

Contenido Descargable

Descargar PDF
https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/f7623h04j

Descriptions

Attribute NameValues
Creator
Abstract
  • The outsourcing of data storage and related infrastructure to third-party services in the cloud is a trend that has gained considerable momentum in the last decade due to the savings it affords companies in both capital and operational costs. Although encryption can alleviate some of the privacy concerns associated with cloud storage, it comes at the cost of decreased utility of data once it is in the cloud. For instance, cloud services for searching over a large set of files are useless when those files are encrypted using standard, randomized techniques. Moreover, even though the files are encrypted, the historical access pattern over a data set (e.g., file access times/frequency, memory read/write locations) can leak significant information to malicious parties that can be combined with other metadata to partially reveal the file contents. Two research thrusts that address these issues are Symmetric Searchable Encryption (SSE) and Oblivious RAM (ORAM). SSE schemes provide an efficient means for searching over encrypted data that is stored on an untrusted server. These schemes, however, are inherently susceptible to statistical attacks by observing the history of de- terministically encrypted search queries, along with the access pattern resulting from those queries. ORAM, by design, can prevents such leakages, but it introduces high communication overhead, among other challenges. In this work we introduce new ORAM schemes tailored for more efficient private access to tree data structures, which are commonly used as search indexes. We make multi-faceted contributions that include: (i) a formal definition for obliviousness in the relaxed setting where the external storage provider knows we are accessing a tree data structure, (ii) a formal proof showing that our schemes satisfy this definition, with an- alytical and empirical results showing that we reduce transmission overhead by several factors over state-of-the-art ORAM schemes, and (iii) a Python software package called PyORAM (Python-based Oblivious RAM) that provides researchers with a powerful set of tools to build upon current ORAM methods tailored for the cloud storage set- ting. Our work has broad impacts for searchable encryption via private database search with reduced transmission requirements, which is vital for privacy-critical applications in governmental and healthcare systems.
License
Resource Type
Fecha Disponible
Fecha de Emisión
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Academic Affiliation
Non-Academic Affiliation
Subject
Declaración de derechos
Publisher
Peer Reviewed
Language
Replaces

Relaciones

Parents:

This work has no parents.

En Collection:

Elementos