Moving personal data into the Cloud (e.g., dropbox, google drive, iCloud, etc.) is an increasing trend in recent years. This is because Clouds services guarantee high reliability, anywhere/anytime accessibility of our data and other convenient functionalities.

 As everything in life, this convenience comes at a price: once our data is moved into the cloud we lose control over it, in fact, we implicitly delegate the protection of our privacy to the Cloud servers. This is highly risky: the repeated Cloud breaches — that already exposed private data of million of users — proves that as long as our valuable information is stored in the Cloud, chances are that it will be illegitimate accessed.

 This brings up the question: can we enjoy Cloud services while still maintaining full control over our data? This question is the starting point of my talk. First, I will discuss the new challenging requirements posed by private cloud computation — especially when the computation involves big data— and how traditional approaches for guaranteeing privacy fail in addressing them, motivating the need of completely new techniques. I will then show our new, theoretical, approach, appeared in STOC 2015, for enabling privacy in cloud computation over big datasets. I will conclude the talk discussing how the road towards incorporating privacy in cloud computation is long and will require combined effort from both the theoretical and the applied community.

Alessandra Scafuro is a Post-Doctoral researcher in the Department of Computer Science at Boston University and Northeastern University. Her research is in the area of cryptography and data privacy with focus on the theoretical foundations of secure computation. She was a research fellow at Simons Institute for the Theory of Computing (UC Berkeley), and held a Post-doctoral appointment at UCLA from 2013 to 2014. She received her PhD in Computer Science from University of Salerno, Italy, in 2013. Her thesis on “Secure Computation Under Network and Physical Attacks” was rewarded with the “Best PhD Thesis in Theoretical Computer Science Award 2013”, from the European Association for Theoretical Computer Science (Italian Chapter).