Seyed K. Fayaz

I have received a Ph.D. in Electrical and Computer Engineering from Carnegie Mellon University, where I was part of CyLab Security and Privacy Institute. I was supported by a VMware Graduate Fellowship, CMU Bertucci Fellowship, and Renaissance Technologies Fellowship.

Research (Google Scholar)

My Ph.D. research, under the supervision of Professor Vyas Sekar, was about network security with a focus on correct policy enforcement. This involves two questions:

  1. How to make network policy enforcement more flexible and less error-prone to begin with? To provide secure and high-performance connectivity, network operators need to enforce a variety of policies (e.g., host A cannot talk to host B, traffic going to department C must first go through a firewall and then a proxy). Enforcing such policies is complex and error-prone. My goal here has been to dramatically simplify network policy enforcement via Software-Defined Networking (SDN) and Network Functions Virtualization (NFV).
  2. How to check whether the network is actually doing the right thing? Even with careful policy enforcement, network bugs may occur (e.g., network outages due to misconfigured routers, or security breaches due to traffic bypassing firewalls) costing organizations money and reputation. The following work takes a step forward in systematically reasoning about the behavior of networks involving both traditional network devices (i.e., switches and routers) as we all as advanced appliances (e.g., firewalls, IDS/IPS, NATs, and proxies). As an aside, I also had a chance to look into Information-Centric Networking (ICN) to determine whether pervasive caching, which is one of the core tenets of ICN, is worth it. (The answer is no.)

Honors and Awards



Professional Service

Research Collaborators

During my Ph.D., I was exceptionally fortunate to learn from a fantastic group of bright researchers, inspiring engineers, and passionate entrepreneurs: Michael Bailey (UIUC), Sagar Chaki (CMU), Luis Chiang (DT Labs), Michael Collins (Redjack), Ari Fogel (Intentionet), Matt Fredrikson (CMU), Ali Ghodsi (Databricks), Victor Heorhiadi (UNC), Mohammad Ali Kaafar (CSIRO), Teemu Koponen (Styra), Yin Lin (VMware), Bruce Maggs (Duke/Akamai), Ratul Mahajan (Intentionet), Todd Millstein (UCLA), Jeff Mogul (Google), Michael Reiter (UNC), Srini Seshan (CMU), Scott Shenker (UC Berkeley), Yi Sun (Chinese Academy of Sciences), Yoshiaki Tobioka (NTT Communications), Amin Tootoonchian (Intel), Steve Uhlig (QMUL), George Varghese (UCLA), and Minlan Yu (Yale).