[cs-talks] Upcoming CS Seminars: NRG (Monday) + PhD Proposal (Tues)
fgreen1 at bu.edu
Mon Sep 28 10:13:02 EDT 2015
Central Control Over Distributed Routing
Presenter: Cody Doucette, BU
Authors: Stefano Vissicchio, Olivier Tilmans, Laurent Vanbever, Jennifer Rexford
Monday, September 28, 2015 at 11am in MCS 148
Abstract: Centralizing routing decisions offers tremendous flexibility, but sacrifices the robustness of distributed protocols. In this talk, I will present Fibbing, an architecture that achieves both flexibility and robustness through central control over distributed routing. Fibbing introduces fake nodes and links into an underlying link-state routing protocol, so that routers compute their own forwarding tables based on the augmented topology. Fibbing is expressive, and readily supports flexible load balancing, traffic engineering, and backup routes. Based on high-level forwarding requirements, the Fibbing controller computes a compact augmented topology and injects the fake components through standard routing-protocol messages. Fibbing works with any unmodified routers speaking OSPF. Experiments also show that it can scale to large networks with many forwarding requirements, introduces minimal overhead, and quickly reacts to network and controller failures.
Paper link: http://conferences.sigcomm.org/sigcomm/2015/pdf/papers/p43.pdf
How I Know You Printed My Email
Alexander Pretschner, Technische Universität München, Germany
Wednesday, September 30, 2015 at 10:00AM
Abstract: This talk tackles the problem of specifying, monitoring and enforcing data usage requirements of the kind, “print my email at most twice,” “notify me upon dissemination of my address,” “no more than three copies of a confidential document in the company,” “delete all copies of a movie within thirty days,” “keep financial record for five years,” and the like. We first sketch how to formally express such requirements, and how to semi-automatically transform user-level policies into technical policies that can be observed or enforced by our infrastructure. As a second step, we present this infrastructure that can act both post factum, for accountability purposes, and preventively. It builds on two main ideas. One, requirements come at various levels of abstraction: prohibiting screenshots, writing files, playing songs, copying database rows can most conveniently observed and controlled by monitors at different layers of abstraction. Two, when data is to be protected, usually all of its representations are meant to be protected: a picture comes as network packets, pix map, cache file, DOM object. This requires information flow tracking technology across the layers of a system and across systems. Practical information flow tracking across layers and across systems with multiple monitors faces significant challenges. These include performance, over-approximations as well as completeness of the data flow analyses, and security of the infrastructure. Time permitting, we sketch mitigations based on quantity and structure of data as well as hybrid static-dynamic information flow trackers. Bio: Alexander Pretschner currently is a visiting researcher in MIT’s distributed information group. Since 2012, he has held the chair of software engineering at Technische Universitat Muenchen in Germany. Prior appointments include positions at the Karlsruhe Institute of Technology, the Fraunhofer Institute for Experimental Software Engineering, and ETH Zurich. Research interests include systems testing, security, and data usage control. More information at https://www22.in.tum.de/en/pretschner/.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the cs-talks