SDN Summit

Mon Nov 10 05:10:00 EST 2014 Mon Nov 10 05:10:00 EST 2014, Mon Nov 10 05:10:00 EST 2014 - Mon Nov 10 23:10:00 EST 2014 Mon Nov 10 23:10:00 EST 2014, Mon Nov 10 23:10:00 EST 2014

New York City, NY, 

Contact: Jia Wang

Short Description of SDN Summit


University of Tel Aviv: Yehuda Afek EE, Technion, M.Sc and Ph.D. in CS, UCLA 1983 and 1985. Member of Technical Staff at AT&T Bell-Labs 1985-88. Faculty of CS in Tel-Aviv University 1988-present. In 2001 he co-founded Riverhead Networks, developed the DDoS Guard enabling clean pipes mitigation of Distributed Denial of Service attacks on the Internet, keeping several Fortune-5 and 500 companies in the air despite being attacked. Worked with tier one providers such as AT&T to start and design clean pipes managed service. Riverhead was acquired by Cisco in 2004, Director of Technology in Cisco 2004-2009. Head of the Blavatnik School of Computer Science Tel-Aviv University


Stanford University: Mohammad Alizadeh
Mohammad Alizadeh received his Ph.D. in Electrical Engineering from Stanford University in 2013. Starting next year, he will join the EECS department at MIT as an assistant professor. He is currently a principal engineer at Cisco, which he joined through the acquisition of Insieme Networks in 2013. His research interests are broadly in networked systems, with focus on datacenter networking and cloud computing. In his dissertation work, he developed high performance packet transport mechanisms for datacenter networks. His research has garnered significant industry interest: the Data Center TCP (DCTCP) congestion control algorithm has been integrated into the Windows Server 2012 operating system; the QCN algorithm has been standardized as the IEEE 802.1Qau standard; and most recently, the CONGA adaptive network load balancing mechanism has been implemented in Cisco's new flagship Application Centric Infrastructure products.


AT&T: Al Blackburn
Interests: SDN, NFV, UDNC / D2 architecture


AT&T: Paul Bartoli
I lead a team of engineers focused on defining the architecture for instantiating, managing and monitoring the virtual environment (SDN plus virtual infrastructure) using realtime control mechanisms. ECOMP - The management framework to configure, monitor and manage the virtual environment in a fully automated fashion.


AT&T: Vimal Begwani I joined AT&T in 1986. For last several years I have been leading work on Service Assurance Target Architecture for AT&T's wireless and wireline network and services. To support AT&T's transition to Domain 2.0, I have been given responsibility to lead architecture for Data Collection, Analytics, and Event (DCAE) and Policy Management. These two critical components in overall ECOMP (Enhanced Control, Orchestration, Management and Policy) platform to support real-time closed-loop optimization. Policy platform and Real time data collection and anlytics to support real-time closed loop optimization of virtualized network and service functions.


Duke University: Theophilus Benson Theophiilus Benson is an Assistant Professor in the Computer Science Department at Duke Univeristy. His research interests include solving practical networking and distributed systems problems, with a focus on improving the reliability and performance and an emphasis on Software Defined Networking, data centers, clouds, and configuration management. His works has earned him the following awards: a Yahoo! Academic Career Award and best paper at IMC (Data center measurements). He received a Ph.D from University of Wisconsin, Madison. Before joining Duke University, he was a Post-Doc at Princeton University. " "My work aims to reduce complexity in network management and in doing so improve reliability, performance and security of networks. To address this complexity, I developed principled techniques to model, capture and understand the interactions between systems and often ignored external factors. Then, I leverage industry collaborators to validate my systems against large-scale deployments, thus allowing me to enrich my techniques to tackle the unexpected issues that appear in the wild. For example, in my work on network management I analyzed design choices for configuring network protocols, modeled interactions between protocol configurations and operator expertise, and evaluated my models on several networks including AT&T’s. This approach to research results in unique techniques that have the following far- reaching implications: for proactively detecting network outages; for designing novel data center architectures cluster schedulers; and for deploying intricate applications within the cloud. My work combines techniques from networking, theory, and programming languages to improve the security, reliability and efficiency of networks. For example, my current work on SDN controllers borrows symbolic execution techniques from programming languages to detect and eliminate fault inducing network events.


AT&T: Lee Breslau Lee Breslau is a Distinguished Inventive Scientist in the Networking and SQM Research Center in AT&T Labs. He received his PhD from the University of Southern California in 1995. From 1994 until 1999, when he joined AT&T, he was a member of the research staff at the Xerox Palo Alto Research Center. Throughout his career, Lee's research focus has been on Internet architecture and protocols, network measurement, multicast routing, and mobility service quality.


AT&T: Anthony Brigandi Tony Brigandi is a Director in the Optimization, Reliability and Customer Analytics Division of AT&T Labs in Middletown, N.J. Tony holds a Ph.D. from NYU-Poly in Operations Research. He currently leads a team responsible for access optimization, customer focus reliability, and network optimization & SDN. His research interests include mathematical programming, heuristics, and simulation. He is a member of INFORMS, an Edelman Laureate and Franz Edelman First Prize winner. routing with sdn controller.


AT&T: Ioannis Broustis Ioannis is a Senior Inventive Scientist at AT&T Labs Research. He received his PhD from the Department of Computer Science & Engineering at the University of California, Riverside, in 2007. He has been with AT&T Labs Research since August 2012. His interests are in the area of Service Quality Management for UDNC. Prior to joining AT&T, Ioannis was a Member of Technical Staff at Alcatel-Lucent, USA, where he worked on security standards for network and middleware systems, as well as on end-to-end security solutions for multimedia over IP and secure cloud services. From February to December 2009, and from July to September 2008 he was a visiting researcher at UC Riverside in the Networks and Communications Laboratories, where he worked on smartphone and network security, as well as wireless network coding. From January 2008 to February 2009 he was a researcher at CERTH, working on developing network coding frameworks and testbed federations. From August 2008 to January 2009 he was an adjunct assistant professor in the Department of Computer & Communications Engineering at the University of Thessaly, Greece. From July 2006 to December 2006 he was an intern at Intel Research in Cambridge, UK, where he worked on Self Organized Wireless Access Networks. During summer 2005 he was an intern at Nokia R&D in Boston, MA, working on performance evaluation techniques for mobile Internet browsers. During summer 2001 he was an intern at Cosmote S.A in Athens, Greece, working on coverage optimization procedures and tools for cellular networks. "Closed-loop Service Quality Management for Domain 2.0; prototyping and experimentation/measurements on the efficiency and practicality of automated network troubleshooting in SDN and NFV-based broadband networks and services.


AT&T: Cagatay Buyukkoc Cagatay have BS and MS degrees from METU, Ankara, Turkey and a PhD degree in EECS from University of California, Berkeley. His area of specialization was on Networking and Stochastic Control. He also completed an Executive MBA program with Wharton extension while working at Bell Labs.
AT&T:  Created a broad resiliency framework that would help a rethinking as part of the design strategy (the network/system) to provide and maintain an acceptable level of services in the face of various faults and challenges to normal operation. (CTD). This framework also introduces a Resiliency manager with an E2E view that can help track per application resiliency state space and take re-mediadtion/recovery actions as needed. 
Worked on various target architecture development of Fixed Mobile Convergence in the areas of Policy, QoS (co-created the TA), and Applications. Conceptualized, created and introduced to marketing work “Cost of QCI” concept that is being used by marketing and strategy groups.
Led the development of the QoS for WebRTC strategy. This is a game changer area that would also benefit from SDN/NFV frameworks. 
Created and lead the Congestion Indication Target Architecture. 

Other companies (several start-ups and vendors):

Worked in several management and research positions:
Led the development of transport optimization tools (e.g., application, transport, optical and physical layers).
Created the optical ring design tool, this tool was used extensively in other vendor products
Worked in ITU-T and led the QoS group
Led the broadband/Ethernet access strategy, folded that into aggregation, optical networks and, NGN architectures;
Conceptualized and led other projects in DWDM systems, Data storage and retention solutions, as well as, intellectual property, innovations and service introduction

Joined Lucent in a new division (under Rich Gitlin), ComSoft group, that was tasked with  developing software for communication networks, where he helped build Congestion Management and Measurement Based Control algorithms.  He also worked in a group developing company strategy in New Technology and Service Development.


Old AT&T:
He joined AT&T Bell Labs Performance analysis department and was responsible for creating performance models for different systems and architectures. He created many service and product models based on IP and ATM technologies and also helped build a Lab that tested and measured new ideas introduced that involve multicasting, video distribution, real time measurement methodologies and caching (He holds AT&T’s early patent on Caching based on geographical information even predating Akamai’s).
"          "The best definition of architecture (imo) is “constraints that deconstrain (CTD)”; in other words the set of constraints (SDN) that you put between the set of applications and on the set of diverse hardware that would allow the biggest benefit for the end users and AT&T. Since we are moving towards a new economy where deployment and amortization cycles are only 2-3 years before everything goes to a minimal marginal cost regime, the frameworks should be very flexible, scalable, programmable and completely auto-managed E2E (using COTS as much as possible is an economic necessity (CTD), not the theme of the day) not in specific islands. SDN/NFV work is a great opportunity in this regard.
AT&T network has many constraints that should be included in the designs (CTD), and new services introduction environment should be able to utilize AT&T unique value adds (e.g., spectrum, mobility, reach, big data, security (AAA), CDN etc.) to extract the most value, in a flexible and resilient way in virtualized environments. SDN/NFV frameworks provide some great tools however; lots of work remains, especially in the understanding of complexity vs robustness considerations. ISO layering needs a complete rethinking along the lines of Information Theory and separation theorem.

In addition, I am leading a group within IEEE SDN framework initiative (preindustrial track committee). The purpose is to help create technical and economic conditions of interoperability and feasibility of technologies around SDN/NFV frameworks (widened to include the RAN) within the industry and academia.  In order to do this the committee is analyzing and evaluating various potential methodologies and will make recommendations and create industrial agreements, and help other bodies (e.g., ETSI). This committee is also looking at opportunities for POC work and establishes relationships with other groups doing relevant work in Europe, Asia, Americas, Africa, etc. to increase collaboration & reduce duplication.


UIUC: Matt Caeser Matthew Caesar is an associate professor in the Department of Computer Science at the University of Illinois at Urbana-Champaign. His research involves improving the reliability and security of distributed systems and networks with an emphasis on software-defined networks, data centers, and the Internet. His research has led to several commercial products and deployed systems. He received the NSF CAREER Award (2011), DARPA CSSG membership (2011), and is a CAS Fellow (2013).


University of Pennsylvania: Chen Chen My name is Chen Chen. I am a fourth-year Ph.D student from University of Pennsylvania. My research interest mainly focuses on networking and formal verification. I collaborate with Prof. Boon Thau Loo (University of Pennsylvania) and Prof. Limin Jia (Carnegie Mellon University) on verification of controller programs in software-defined network. We use declarative programming language (i.e., Network Datalog) to specify programs, and develop a sound proof system for verifying properties written in first-order logic, based on the operational semantics of the language. The properties we verify include but not limited to: security in inter-domain routing protocols, correct reactive flow installation of the controller and correct network configuration (e.g., loop-freeness). I also work closely with AT&T research lab, to develop useful cloud techniques based on software-defined networking. My collaboration with researcher Changbin Liu aims to build a large-scale software-defined layer-2 network. Our system is scalable, efficient and flexible, spanning multiple datacenters while allowing the network administrator to spawn new virtual machines quickly and migrate them seamlessly.


AT&T: Margaret T Chiosi DMTS focus on SDN and Virtualization technology AT&T SDN Realization


AT&T: Angela Chiu Angela L. Chiu received her S.M and Ph.D. degrees, both in electrical engineering and computer science from MIT, Cambridge, MA, in 1993 and 1997, respectively. She was a research staff member in the Optical Communication Technology group of MIT Lincoln Laboratory from April to November 1997. In November 1997, she joined AT&T Labs and lead architecture and design on IP QoS, MPLS, traffic engineering, restoration and control plan design for IP over Optical architecture. Her team pioneered the first IP class of service for AT&T Worldnet service. Now she is a principal Inventive Scientist at Network Evolution Research Department. The planning tool for AT&T core transport network called Hawkeye was invented and prototyped by the research team that she has been leading. During 2001 and 2002, she was a principal architect at Celion Networks, an optical start-up. Dr. Chiu has published extensively in many technical journals, conferences, and technical book, and has served as conference and program chairs for numerous conferences. She holds twenty U.S. patents. She is a co-author of 6 IETF standards. Her research interests include architecture, planning, design, and analysis of IP, Ethernet, and optical networks, software-defined networks, network function virtualization. My research interests include architecture, planning, design, analysis, management of IP, Ethernet, and optical networks, software-defined networks, network function virtualization. She is leading a research prototype tool called Birdseye which provides visualization and analysis for IP planning, technology and operation folks to move the network toward SDN-enabled IP core network. She is also a key member of the Next-Gen Router Farm project team which aims to revolutionize the IP network edge with flexible, reliable, scalable, cost-effective, and manageable SDN-based edge platform.


AT&T: Gustavo De Los Reyes Lead AT&T Security R&D Security research in Cloud, Mobility, Big Data


AT&T: Robert Doverspike Robert Doverspike received his undergraduate degree from the University of Colorado and Masters and Ph.D. degrees from Rensselaer Polytechnic Institute (RPI). He began his career with Bell Labs and, upon divestiture of the Bell System, went to Bellcore (later called Telcordia). He returned to AT&T Labs (Research) where he is now Executive Director of Network Evolution Research. Dr. Doverspike has made extensive contributions to the field of optimization of multi-layered transmission and switching networks and pioneered the concept of packet transport in metro and long distance networks. He also pioneered work in spearheading the deployment of new architectures for transport and IP networks, network restoration, and integrated network management of IP-over-optical-layer networks and Software Defined Networking. He has over 1500 citations to his books and articles over diverse areas/publications such as Telecommunications, Optical Networking, Mathematical Programming, IEEE Magazine, IEEE Communications Society, Operations Research, Applied Probability, and Network Management. Dr. Doverspike holds many professional leadership positions and awards, such as INFORMS Fellow, IEEE Fellow, member of Optical Society of America (OSA), co-founder of the INFORMS Technical Section on Telecommunications, OFC Steering Committee, DRCN Steering Committee, and Associated Editor for JOCN (Journal of Optical Communications and Networking). SDN in carrier networks, bandwidth-on-demand, NFV for PE-routers and CE_routers and edge of packet network, automation of planning and provisioning in metro networks.


Georgia Tech: Nick Feamster  Nick Feamster is a professor in the College of Computing at Georgia Tech. He received his Ph.D. in Computer science from MIT in 2005, and his S.B. and M.Eng. degrees in Electrical Engineering and Computer Science from MIT in 2000 and 2001, respectively. His research focuses on many aspects of computer networking and networked systems, with a focus on network operations, network security, and censorship-resistant communication systems. In December 2008, he received the Presidential Early Career Award for Scientists and Engineers (PECASE) for his contributions to cybersecurity, notably spam filtering. His honors include the Technology Review 35 "Top Young Innovators Under 35" award, the ACM SIGCOMM Rising Star Award, a Sloan Research Fellowship, the NSF CAREER award, the IBM Faculty Fellowship, the IRTF Applied Networking Research Prize, and award papers at the SIGCOMM Internet Measurement Conference (measuring Web performance bottlenecks), SIGCOMM (network-level behavior of spammers), the NSDI conference (fault detection in router configuration), Usenix Security (circumventing web censorship using Infranet), and Usenix Security (web cookie analysis).        "My research focuses on developing techniques, algorithms, and protocols that make the network easier to man- age, more secure, and more available. I am an experimental networked systems and security researcher. A hallmark of my work is building, designing, and monitoring real networked systems.

Nobody notices when the network works well, but everyone suffers when it doesn’t. Thus, communications networks should be both secure and available. Network security has many facets, ranging from the ability to stop “unwanted traffic” (e.g., spam and denial-of-service attacks) to the ability to trace back attacks to their perpetrators (“account- ability”). Availability means that the network must provide good performance for users whenever they want to use it—unfortunately, the increasing complexity of the network, coupled with hardware faults, software bugs, misconfig- urations, and malice, make it difficult to achieve this goal. Unfortunately, these two important goals have also been among the most difficult to achieve. Breakthroughs require not only extensive domain knowledge, but also the ability to apply techniques from a wide range of areas, ranging from economics to machine learning. My work combines domain knowledge, extensive interactions with network operators, techniques from a wide range of disciplines, and— perhaps most importantly—the competence and tenacity to implement and deploy these systems in practice. This unique combination has allowed me to build one of the few networking research groups in the world that interacts directly with network operators to deploy fundamentally new systems and technologies in real-world networks.

I discover interesting and challenging practical problems through frequent discussions and meetings with network operators and people in industry. I then tackle these problems from first principles, develop new methods, and transfer these solutions back to practice in the form of working systems. I have tackled a variety of problems in network operations, ranging from real-time network diagnosis to stopping unwanted traffic like spam to architectures for fast failure recovery. Many people—most notably, operators “in the trenches”—are also working on these problems. Unfortunately, many of the people who have the domain knowledge that best equip them to solve these problems are busy with day-to-day operations, putting out fires as they arise but rarely taking time to think about fundamental changes to the network that might eradicate these problems. My research fills this niche. I first devise methods to understand the nature of the problem in practice. I tackle domain-specific problems with tools and techniques from other disciplines—ranging from machine learning to economics to program analysis—whose principles might provide insights into a new, previously undiscovered solution. I then devise a new approach or solution, and I transfer it to practice through implementation and deployment of real-world systems.

My research in this broad area is currently focusing on several themes: (1) Internet censorship and open access; (2) home and access networks; and (3) software defined networking. These themes, which I have been developing in the past several years since receiving tenure, build on the broader research themes I have developed on network security and operations. I first survey the new leadership roles that I have assumed in research, teaching, and service. Then, I discuss each of the new research themes I have developed since tenure and the impact that they have had on both other researchers and on industry.


Brown University: Rodrigo Fonseca I am an assistant professor at Brown University's Computer Science Department. I do research in computer systems, specifically in Networking, Distributed Systems, and Operating Systems. I currently have three main lines of investigation: software-defined networking, understanding the execution of large scale distributed systems, and energy management for mobile devices. There are two overarching themes that tie these together: first, I seek to understand the dynamic runtime behavior of distributed and/or complex systems, and second, how information from the end-users of these systems can guide execution parameters across several layers of abstractions. In the realm of networking, I am interested in how SDNs can enable a much tighter integration between the network infrastructure and the end-user applications running on the network. One example of this is our work on Participatory Networking (Sigcomm’13), which provides an API to the network that applications can use to provide hints about future usage and request service properties including bandwidth, latency, and security.


Cornell University: Nate Foster Nate Foster is an Assistant Professor of Computer Science at Cornell University. His research focuses on developing language abstractions and tools for building reliable systems. Some topics of interest include semantics, type systems, bidirectional languages, provenance, data synchronization, and mechanized proof. His recent work has focused on developing domain-specific languages for software-defined networks as a part of the Frenetic project. Nate received a PhD in Computer Science from the University of Pennsylvania in 2009, an MPhil in History and Philosophy of Science from Cambridge University in 2008, and a BA in Computer Science from Williams College in 2001. He was a postdoc at Princeton University from 2009-2010. His awards include a Sloan Research Fellowship, an NSF CAREER Award, a Yahoo! Academic Career Enhancement Award, and the Morris and Dorothy Rubinoff Award.


AT&T: FRANCIS, MARK Mark Francis is Vice President, AT&T Technology Operations Planning. Mark's responsibilities include Technology Operations Planning for all AT&T Technology Infrastructure worldwide, including networks and compute/storage/OS. Mark also supports Corporate Business Continuity Planning and Contingency Planning for the corporation. In addition, Mark and his team support Technology Operations in the ingestion of new services and Technologies (e.g. VoLTE, Digital Life, Small Cells, NetBond, Cloud-based Applications) and an intense day-to-day focus on Operations Optimization via Automation and Process enhancements, including End-to-End Service Quality Management which looks across the entire eco-system that delivers services to Customers and automates the understanding of service health, quality, recovery priority, and targets issues to their root cause for fast resolution. Finally, Mark provides strategic planning on how the Operations Model will evolve over time. Interests: SQM, SDN


AT&T: I received my BA in Computer Science and Technology from Tsinghua University, China, my MS in Computer Science from Boston University and my PhD in Computer Science from University of Massachusetts at Amherst. My PhD dissertation was on investigating distributed information dissemination and retrieval mechanisms through multicasting, peer-peer networks and publish/subscribe systems. Since November 2013, I have been a lead member of technical staff at AT&T Labs -Research. Before that, I had worked as a senior postdoctoral researcher in the Advanced Computer Networks Research Group of Umass Amherst after my PhD, a senior member of technical staff at AT&T Labs - Research between February 2003 and February 2007, the director of search engine research at Adverplex, Inc. from February 2007 to May 2008, and a principal member of technical staff at AT&T Labs - Research from June 2008 to November 2013. Research Interest: Service quality management, network management, anomaly detection, data mining.


Stony Brook University: Phillipa Gill Phillipa Gill is an assistant professor of computer science at Stony Brook University (SUNY). Prior to joining Stony Brook she spent a year working with the Citizen Lab at the Munk School of Global Affairs, University of Toronto. She completed her Ph.D. at the University of Toronto in 2012 and holds an M.Sc. and B.Sc. in computer science from the University of Calgary. During her Ph.D., she spent time as a visiting researcher at AT&T Labs–Research, Boston University, and Microsoft Research. Phillipa’s research focuses on network measurement and characterization. She uses measurement, data analysis, and ideas from economics to improve the security and reliability of network. Recently this has led her to work on topics related to online information controls and net neutrality. Her past work includes studying deployability of new secure routing protocols in ISP networks and reliability of data center networks. She is generally interested in addressing network management challenges that can improve the security and reliability of carrier networks.


AT&T: Vijay Gopalakrishnan Vijay Gopalakrishnan is a Director in the Network Evolution Department in AT&T Labs – Research, leading a small team focused on network architecture, protocols, policy, and management. Vijay is broadly interested in systems challenges in networking. He has worked on innovative solutions in the space of network management, content delivery, and the mobile web and was awarded the AT&T CTO award in 2009 for his work on using SDN to efficiently manage network edges. Prior to joining AT&T in 2006, Vijay got his Masters and PhD in computer science from the University of Maryland, College Park. He is a member of ACM and IEEE. I lead a small team that is currently working on multiple projects using SDN and NFV in the carrier setting. Our goal is to use these capabilities to not only design and build next generation carrier networks, but also build techonlogy that allows easier operation and management


University of Michigan: Yihua Guo My name is Yihua Guo. I am a third-year PhD student in Computer Science and Engineering Division at University of Michigan. My research interests are end-to-end performance of cellular networks and SDN support for cellular infrastructure. I’m working with my advisor, Professor Zhuoqing Morley Mao, on identifying problems and improving performance of network protocols on smartphones over cellular networks. We are also designing a handover mechanism in the SDN-based cellular infrastructure to provide better handover support and performance. Up to now, I co-author the 2013 SIGCOMM paper titled “An In-depth Study of LTE: Effect of Network Protocol and Application Behavior on Performance”, where I did controlled local experiments and network trace analysis to confirm the problems we found in an extensive data trace collected at a commercial LTE network.


Georgia Tech: Arpit Gupta Arpit Gupta's research focuses on intersection of SDN and Internet routing. His project on software defined IXPs (SDXs) tries to bring the merits of Software Defined Networking (SDN) to Internet routing. SDX develops new programming abstractions that allow participating networks to create and run SDN applications and a runtime that both behaves correctly when interacting with BGP and ensures that the SDN applications do not interfere with each other. It also ensures that the system scales, both in rule-table size and computational overhead. Arpit Gupta is currently a second year PhD student at Georgia Tech. At Georgia Tech he works under supervision of Dr. Nick Feamster and is also guided by Dr. Jennifer Rexford, Princeton for his SDX project. Before joining Georgia Tech, he completed his masters from NC State University and undergrad from Indian Institute of Technology, Roorkee, India. Webpage:


AT&T: Bo Han Bo Han received the Bachelor’s degree in Computer Science and Technology from Tsinghua University in 2000, the M.Phil. degree in Computer Science from City University of Hong Kong in 2006 and the Ph.D. degree in Computer Science from the University of Maryland in 2012. He is currently a senior inventive scientist at AT&T Labs Research. His research interests are in the areas of wireless networking, mobile computing, software defined networking and network functions virtualization, with a focus on developing simple yet efficient and elegant solutions for real-world networking and systems problems.SDN + NFV: VPN on demand and Next Generation Router Farm


AT&T: Shuai Hao Shuai joined AT&T Labs in July 2014, right after completing his Ph.D. at USC. His research interests are in the area of networked systems, mobile and cloud computing. In the past, he has worked on mobile app energy profiling and frameworks for app instrumentation and testing. Networked systems. Mobile and cloud computing


AT&T: Pramod A. Jamkhedkar Ph.D., Computer Engineering for Domain 2.0; prototyping and experimentation/measurements on the efficiency and practicality of automated network troubleshooting in SDN and NFV-based broadband networks and services.


USC: Masoud Moshref Javadi Masoud Moshref is a 5th year PhD candidate in University of Southern California. He works on resource virtualization in Software-Defined Networks in Networked Systems Lab under supervision of Ramesh Govindan and Minlan Yu. He got MSc and BSc in Information Technology Engineering from Sharif University of Technology in Iran. Software defined networking centralizes control plane functionality, separating it from the data plane which is responsible for packet forwarding. Many management tasks such as finding heavy hitters for multi-path routing may run using SDN in a network with limited resources. However, by abstracting them from resources at individual switches, a resource manager at the controller can optimize their resource usage. As management tasks often have a measurement-control loop, my projects, DREAM and vCRIB, work on measurement and control tasks, respectively: First, Dream ensures a minimum user-specified level of accuracy for tasks instead of allocating a fixed amount of resources to each task. Therefore, it dynamically allocates resources across tasks in reaction to traffic dynamics and task dynamics, which allows resource multiplexing. DREAM is 2x better at the tail of minimum accuracy satisfaction comparing to current practice even in cases with moderate load. Next, vCRIB automatically distributes control rules on all switches in the network giving the abstraction of a centralized rule repository with resources equal to the combined resources of all switches. vCRIB can find feasible rule placement with less than 10% traffic overhead in cases where traffic-optimal rule placement is not feasible with respect to CPU and memory constraints.


University of Minnesota: Cheng Jin Cheng Jin is a third­year Ph.D. student in Computer Science at the University of Minnesota, advised by Prof. Zhi­Li Zhang. She received her B.Eng. degree in Network Engineering from the Dalian University of Technology in China in 2012. Cheng's primary interests are networking and systems. Her current research focuses primarily on data center networking, software defined networks, and cloud storage systems. Cheng has been working on several cloud service related research projects. In one of her first research projects, Cheng conducted extensive network measurements to study the performance of Dropbox and other cloud­based file sync systems such as Google Drive and One Drive, with the goal to understand how these cloud storage services conduct data synchronizations and their impact on network traffic usage. These research efforts have led two research papers appeared in Middleware’13 and IMC’14 respectively. Since summer 2013, Cheng has been collaborating with a couple of colleagues from AT&T Labs ­­ Research to analyze security group configuration problems in multi­tenant cloud data centers. A short paper describing the initial research results, entitled “Secgras: Security Group Analysis As a Cloud Service” has been accepted by the IEEE ICNP’14 as a concise paper.


Princeton: Xin Jin I am a fourth-year PhD candidate at the Department of Computer Science, Princeton University. I am working with Professor Jennifer Rexford. I am interested in networking and distributed systems. My current research focuses on software-defined networking (SDN). It consists of three projects, SoftCell, Dionysus, and CoVisor. The SoftCell project rethinks the architecture of cellular core networks. Cellular core networks suffer from inflexible and expensive equipment, as well as from complex control-plane protocols. We design SoftCell, a scalable architecture that supports fine-grained policies for mobile devices in cellular core networks, using commodity switches and servers. The Dionysus project targets the network update problem in software-defined networks. We design, implement and evaluate Dionysus, a system that can perform fast, consistent network updates. Dionysus encodes as a graph the consistency-related dependencies among updates at individual switches, and it then dynamically schedules these updates based on runtime differences in the update speeds of different switches. The CoVisor project explores a new kind of network hypervisor that enables, in a single network, the deployment of multiple control applications written in different programming languages and operating on different controller platforms. It allows multiple controllers to cooperate on managing the same shared traffic. It also abstracts concrete topologies, providing custom virtual topologies in their place, and allows administrators to specify access controls that regulate the packets a given controller may see, modify, monitor, or reroute.


University of Colorado: Murad Kaplan Murad Kaplan is a 3rd year PhD student at the University of Colorado, Boulder working with Professor Eric Keller and spent the summer of 2013 at AT&T working on utilizing modern virtualization and orchestration technology toward the scalability of network functions.


Stanford University: Sachin Katti Sachin Katti is an Assistant Professor of Electrical Engineering and Computer Science at Stanford University and also the Chief Scientist and Co-Founder of Kumu Networks which is commercializing his research on full duplex radios. He received his PhD in EECS from MIT in 2009. His research focuses on designing and building next generation high capacity wireless networks by combining techniques from information and coding theory, RF systems, and networking. His dissertation research focused on redesigning wireless mesh networks with network coding as the central unifying design paradigm. The dissertation won the 2008 ACM Doctoral Dissertation Award - Honorable Mention and the George Sprowls Award for Best Doctoral Dissertation in EECS at MIT. His research has won numerous awards, including the IEEE William Bennett Prize, the Best Student Paper Award at ACM SIGCOMM 2012, USENIX ATC 2013, the Sloan Fellowship, the NSF Career Award as well as Okawa, Hooover, Packard and Terman Faculty Fellowships. " The RAN in the future (LTE-A, 5G) will be dense and heterogeneous. At the same time, traffic demands on the RAN are also scaling and becoming diverse. Scaling capacity and managing such a dense chaotic infrastructure requires tight co-ordination in the infrastructure to tackle interference, as well as the ability to flexibly and dynamically define eNB behavior to handle network and traffic diversity. This research agenda aims to bring a software defined networking approach to designing, deploying and managing the RAN. We show how by systematically decoupling the LTE control and user planes across the entire architecture, from the eNB to the transport, we can build efficient, scalable and tightly coordinated RANs while retaining the flexibility to change the network service definition on the fly. SoftRAN's architecture creates very simple eNBs with standard interfaces and stripped of complex control mechanisms, designs a low latency but practical mobile transport for the control plane, and designs a low latency software controller to implement functions such as CoMP, RAN slicing etc.


USC: Ethan Katz-Bassett I am an assistant professor at USC. I run a networking and systems group with Ramesh Govindan, Minlan Yu, and Wyatt Lloyd. Previously, I worked at Google on mobile web performance. I received my PhD from the University of Washington.

I work on systems to improve Internet reliability and performance, spanning IP, BGP, TCP, content delivery, and the web. I look to providers to find problems. I use network measurements to understand them. Based on the measurements, I design systems to address the problems. Most of my approaches are deployable on today’s Internet. I am particularly interested in the problems that affect dominant Internet players, including cloud providers, large content providers and content delivery networks, and mobile providers. The properties of these networks allow for tailored solutions. To highlight previous projects in topic areas in which I have ongoing work:

Internet Content Delivery: Increasingly, most Internet traffic comes from a small number of content providers, content delivery networks, and cloud providers. We developed techniques to locate all Google servers, as well as the mapping between servers and clients (IMC 2013).

TCP Performance: With Google, we designed new TCP loss recovery mechanisms that resulted in a 23% average decrease in Google client latency (SIGCOMM 2013).

Internet Routing: We run a BGP testbed that connects to real ISPs around the world (HotNets 2014). For example, using the testbed, we built LIFEGUARD, a system to force other networks to reroute around persistent failures (SIGCOMM 2012).

Internet Measurement: Service providers and operators have little visibility into the Internet routes on which they rely. With Reverse Traceroute, we built a system to measure the path taken by an arbitrary destination to reach the user, without control of the destination (NSDI 2010). With PoiRoot, we built a system to pin down the origin of Internet path changes, despite the complex interactions between networks, policies, and protocols (SIGCOMM 2013).


: Eric Keller

Eric.Keller@Colorado.EDU University of Colorado "Eric Keller is an Assistant Professor at the University of Colorado, Boulder. He obtained a PhD from Princeton University in 2011 with a focus on network virtualization and virtualization security. Prior to Princeton, he worked in industry for 7 years at Xilinx, working on topics such as run-time reconfiguration tools and domain specific languages. He joined Colorado after a one year post-doc at the University of Pennsylvania working on cloud security and getting an exposure to software radio. At Colorado, Prof. Keller's group has been exploring all aspect of software-defined networking, from the design of the controller and network virtualization technology, to the use in new environments, to leveraging the network programmability to provide a more active security infrastructure. More information can be found at the group website:


: Adrian J. Kester Adrian is a researcher at AT&T-Labs working on SDN simulation and data analysis and visualization. Starting at Bell Labs in 1977, he has worked in some of its successor companies on various network and business optimization problems. He has a Ph.D. in Statistics from the University of Rochester. Algorithms and performance


AT&T: Nancy K. Kester Nancy has contributed to a variety of functional areas within AT&T, including Network planning and architecture, and network operations, including introduction of High Speed CCS7 signalling links and Segmentation Directory into the AT&T Network. She holds a PhD in Statistics from University of Rochester, where her thesis focused on multivariate data analysis.Optimization functions and SDN controller


AT&T: Thomas J. Killian Tom joined the Computing Science Research Center at Bell Labs in 1982, where he worked on operating systems, networking and espresso machine repair. He has been studying network data as a proxy for customer experience since the mid 1990’s. Service Quality Management on a customer-focused end-to-end basis.