ACM SIGCOMM 2018 Workshops and Tutorials: Monday Early Morning Session

---

Abstract: In this keynote presentation, we introduce our recent research on in-network data analytics and deep machine learning in softwarized infrastructure, especially for emerging new generation mobile networking. Advanced traffic data analytics becomes possible because not only fixed transport equipment but also mobile base stations are getting softwarized facilitating the deployment of complex data processing within network by means of data-plane programmability. In the fifth generation mobile networking (5G), enhanced Mobile Broadband (eMBB) , Ultra Reliable and Low Latency Communication (URLLC) and massive Machine Type Communication (mMTC) are expected to be serviced without interference. While eMBB is planned to be service soon, we posit that URLLC services such as cooperative driving, drone surveillance, etc. become the next focus in near future, and require “mobile network slicing via advanced traffic classification” and “extreme edge computing”. We introduce in-network data analytics and deep machine learning in mobile networking to enable per-application end-to-end fixed mobile network slicing for upcoming 5G infrastructure.

 

Bio: Professor, Dr. Akihiro NAKAO received his B.S. in Physics and M.E. in Information Engineering from the University of Tokyo. He worked at IBM Yamato Laboratory, Tokyo Research Laboratory, and IBM Texas Austin. He received his M.S. and Ph.D. in Computer Science from Princeton University. Since 2005, he has been teaching as a Professor in Applied Computer Science at the Interfaculty Initiative in Information (III) Studies, Graduate School of Interdisciplinary Information Studies (GSII), The University of Tokyo. He has been appointed as Chairperson of Department of GSII since 2016. He has also been appointed Chairman of the 5G Mobile Network Promotion Forum (5GMF) Network Architecture Committee by Japanese government since 2014.

 

---

Abstract: Many of the current DDoS attacks are organised by youngsters, often between 15 and 18 years old. Via booters and other “DDoS as a Service” websites they attack gamers, schools, banks and others. But why have banks and other big organisations so many problems to protect against these attacks? How do these attacks look like, and can scientists help to make the Internet a bit safer? In this talk an overview is given of some current research regarding detection and mitigation of DDoS attacks. Details are given how to measure the number of systems that are under attack, how attribution of such attacks is possible by using fingerprints, and how mitigation techniques like black-holing are being applied.

 

Bio: Aiko Pras is Professor at the University of Twente (UT), the Netherlands, where he is member of the Design and Analysis of Communication Systems (DACS) group. His research interests include Internet security, measurements and management, with a focus on Flow-based measurements, DDoS defence, DNS and anycast. He has chaired the IFIP Technical Committee on "Communications Systems" (IFIP-TC6, 2013-2017), as well as the EU Future Internet cluster (2012-2014). He has been coordinator of the European Network of Excellence on "Management of the Future Internet" (FLAMINGO, 2012-2017), and is currently "Beirat" of the CODE Cybersecurity research institute in München.  He has been Series Editor of IEEE Communications Magazine, Associate Editor of TNSM and JNSM, and is Steering Committee member of many conferences, including IM, NOMS, CNSM, Networking and TMA.

 

---

Abstract: Historically, the conventional wisdom has been that networks should offer fixed functionality for forwarding packets and little else. This appears to be changing, as a new breed of programmable switches match the speed of fixed function devices. This trend raises the question: What services could and should sit in the forwarding plane? In this talk, we explore how a programmable forwarding plane can naturally accelerate the fundamental building blocks of distributed systems. We identify and discuss three key issues for network acceleration: (i) How do we actually program network devices? What are the abstractions and limitations? (ii) What is the right architecture these "networked-distributed systems"? (iii) Given that we are asking the network to do so much more work, how can we be sure that it is implemented correctly? Our results show how we can build provably-correct distributed systems that provide orders-of-magnitude performance improvements over software-based alternatives.

 

Bio: Robert Soulé is an assistant professor at the Università della Svizzera italiana (USI) and a Research Scientist at Barefoot Networks. His research interests are in distributed systems, networking, and applied programming languages. Prior to joining USI, he was a postdoctoral associate at Cornell University. He received his PhD from New York University in 2012, and his BA from Brown University in 1999. For two years, he was a research co-op in the Data Intensive Systems and Analytics Group at IBM T. J. Watson Research Center. He is the recipient of an IBM Invention Plateau Award, a Google Faculty Research Award, and Best Paper Awards at ACM DEBS 2012 and USENIX NSDI 2018.