Matei Zaharia: 2017 Security Workshop


Monday, April 10, 2017
Location: McCaw Hall, Arrillaga Alumni Center

"Splinter: Practical Private Queries for Internet Services"



Many online services let users query public datasets such as maps, flight prices, or restaurant reviews. Unfortunately, the queries to these services reveal highly sensitive information that can compromise users' privacy. To address this problem, we have developed Splinter, a system that protects users' queries on public data and scales to realistic applications. A user splits her query into multiple parts and sends each part to a different provider that holds a copy of the data. As long as any one of the providers is honest and does not collude with the others, the providers cannot determine the query. Splinter uses and extends a new cryptographic primitive called Function Secret Sharing (FSS) that makes it up to an order of magnitude more efficient than prior systems based on Private Information Retrieval and garbled circuits. Our system achieves end-to-end latencies below 1.6 seconds for realistic workloads including a Yelp clone, flight search, and map routing.


Matei Zaharia is an assistant professor of Computer Science working broadly in computer systems. His past work has included datacenter-scale computing systems such as Apache Spark, Spark Streaming and Mesos, resource management algorithms such as Dominant Resource Fairness, and algorithms and systems for large-scale machine learning and genomics. His recent work seeks to combine security with big data by scaling up strong data security techniques. Dr. Zaharia received the 2014 ACM Doctoral Dissertation Award, the VMware Systems Research Award, and best paper awards at NSDI and SIGCOMM in recognition of his research.