Secure Joins with MapReduce - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2018

Secure Joins with MapReduce

Abstract

MapReduce is one of the most popular programming paradigms that allows a user to process Big data sets. Our goal is to add privacy guarantees to the two standard algorithms of join computation for MapReduce: the cascade algorithm and the hypercube algorithm. We assume that the data is externalized in an honest-but-curious server and a user is allowed to query the join result. We design, implement, and prove the security of two approaches: (i) Secure-Private, assuming that the public cloud and the user do not collude, (ii) Collision-Resistant-Secure-Private, which resists to collusions between the public cloud and the user i.e., when the public cloud knows the secret key of the user.
Fichier principal
Vignette du fichier
main.pdf (439.67 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-01903098 , version 1 (16-12-2018)

Identifiers

  • HAL Id : hal-01903098 , version 1

Cite

Xavier Bultel, Radu Ciucanu, Matthieu Giraud, Pascal Lafourcade, Lihua Ye. Secure Joins with MapReduce. FPS 2018 : The 11th International Symposium on Foundations & Practice of Security, Nov 2018, Montreal, Canada. ⟨hal-01903098⟩
446 View
133 Download

Share

Gmail Mastodon Facebook X LinkedIn More