Research‎ > ‎Publications‎ > ‎

Feasibility of a privacy preserving collaborative filtering scheme on the Google App Engine -- a performance case-study

Anirban Basu, Jaideep Vaidya, Theo Dimitrakos and Hiroaki Kikuchi

In: 27th ACM Symposium on Applied Computing (SAC) Cloud Computing track, Trento, Italy.

Year: 2012

Abstract: The cloud is a utility computing infrastructure that has caused a paradigm shift in the way organisations requisition, allocate, and use IT resources. One big challenge is to preserve the confidentiality of information on the cloud. Most typical solutions use cryptographic techniques without considering how well suited they are to the cloud. This paper presents a performance case-study on implementing the building blocks of a privacy preserving collaborative filtering (PPCF) scheme in Java on the Google App Engine (GAE/J) cloud platform. The results show that the GAE/J in its current state exhibits serious performance bottlenecks for the chosen application scenario. This case study highlights the need for better performance from the GAE/J. It also informs the need for validating theoretical cloud security algorithms on real cloud computing platforms in which many performance expectations do not hold.

Fulltext: PDF

Presentation: PDF