Processing analytical queries over encrypted data

Abstract

MONOMI is a system for securely executing analytical workloads over sensitive data on an untrusted database server. MONOMI works by encrypting the entire database and running queries over the encrypted data. MONOMI introduces split client/server query execution, which can execute arbitrarily complex queries over encrypted data, as well as several techniques that improve performance for such workloads, including per-row precomputation, space-efficient encryption, grouped homomorphic addition, and pre-filtering. Since these optimizations are good for some queries but not others, MONOMI introduces a designer for choosing an efficient physical design at the server for a given workload, and a planner to choose an efficient execution plan for a given query at runtime. A prototype of MONOMI running on top of Postgres can execute most of the queries from the TPC-H benchmark with a median overhead of only 1.24× (ranging from 1.03× to 2.33× ) compared to an un-encrypted Postgres database where a compromised server would reveal all data.

TrustedMR: A Trusted MapReduce System Based on Tamper Resistance Hardware

“…Centralized DaaS without secure hardware: Many works [18,22] have addressed the security of outsourced database services (DaaS) by encrypting the data at rest and pushing part of the processing to the server side but none of them can achieve all aspects of security, utility, and performance. In terms of utility and security, the best theoretical solution such as fully homomorphic encryption [12], allows server to compute arbitrary functions over encrypted data without decrypting.…”

“…In terms of utility and security, the best theoretical solution such as fully homomorphic encryption [12], allows server to compute arbitrary functions over encrypted data without decrypting. However, this construction is prohibitively expensive in practice with overhead of 10 9 × [22]. In term of performance, CryptDB [18] provides provable confidentiality by executing SQL queries over encrypted data using a collection of efficient SQL-aware encryptions.…”

“…But this system is not completely secure since it still uses some weak encryptions (e.g., deterministic & order-preserving encryptions [8]). Similarly, MONOMI system [22] securely executes arbitrarily complex queries over sensitive data on an untrusted database server with a median overhead of only 1.24× compared to an un-encrypted database. However, this system still uses some weak encryption schemes (e.g., deterministic encryption) to perform some SQL operations (e.g., Group By, equi-join).…”

“…Privacy violations are legion and arise from negligence, attacks and abusive use and no current server-based approach seems capable of closing the gap 3 . Cryptographic-based solutions have been proposed (e.g., [8,18,22]) to guarantee that data never appear in the clear on the servers but they provide either poor performance, poor security or support a very limited set of computations. Consequently, several attempts of personal data management decentralization have appeared (e.g., [1,3,17]).…”

Order-Preserving Encryption Secure Beyond One-Wayness

“…. , j}), and SQL operations [1,40,41,44]. In cotrast, each of the other primitives mentioned above enables at most one of these operations.…”