Finding the Right Way for High-Performance NFV

Prof. Ryota Kawashima ( Nagoya Institute of Technology, Japan )

Abstract

The trend for softwarizing network functions has reached to the area of core networks where tremendous amount of traffic is flowing on. Fast packet processing frameworks like DPDK are now integral part of NFV-nodes for high-performance networking. However, our study has revealed that further performance advances are necessary even for the cutting-edge NFV-nodes. In this presentation, future research directions in regard to NFV-node performance is given based on our thorough performance analysis results.

Bio.

He received his M.S. degree from Iwate Prefectural University (Japan) in 2007 and also received a Ph.D. degree from The Graduate University for Advanced Studies (SOKENDAI, Japan) in 2010. He has worked as a software engineer at ACCESS CO., LTD. and Stratosphere, Inc. In 2013 he became an assistant professor at Nagoya Institute of Technology (Japan). His research interest is performance aspect of SDN/NFV, and related papers have been accepted for top level journals and conferences, such as IEEE TNSM and IEEE NetSoft. He received the best paper award for 2016 IEICE Communications Society.

Network Intelligence in the age of IoT with SDN and NFV

Dr. Sangho Shin ( Manager, SK Telecom, Korea )

Abstract

In the era of IoT, Big Data, and AI, SKT has been trying to transform to ICT and Data company. As one of the various efforts, we have been gathering data from telco network infrastructure creating data lake system, and we are utilizing the data for operation intelligence through big data analysis.
In this talk, I am going to describe how we collect and analyze the data, focusing on servers and switches in telco cloud infrastructure. We have developed T-CORE for cloud server monitoring and TINA for network monitoring, which collect logs, statistics, and events and store them in TSDB and allow users to analyze the data providing various analysis UIs and tools. Recently, we are working on anomaly detection for recommendation and self-healing, using machine learning and deep learning technologies.
The analysis results are transferred to our control plane in real time or periodically. We have developed SDN based network controller for virtual and physical network, SONA and SONA Fabric, respectively. SONA is the ONOS based carrier grade multi-tenancy virtual network solution, and SONA fabric is the leaf-spine fabric solution. In this talk, I will also explain how effectively we can manage data center network using SONA and SONA Fabric, integrating with T-CORE and TINA.

Bio.

Sangho Shin received B.S from Korea University and received his master’s degree in 2002 and Ph.D in 2008, both from Columbia University in City of New York. He has done research on VoIP on Wireless Network in IRT Lab with Prof. Henning Schulzrinne.
He worked at LG for three years and Samsung for six years as a principal researcher. He also worked at ON.Lab (now ONF) as a visiting scholar and contributed to developing ONOS and Segment Routing use case (CORD Fabric), which was shown at ONS '15. Now, he is working for SK Telecom in Korea and leading the SDN projects at R&D center.
His interests are Wireless Networking, SDN, and Home Networking. He published many papers on top conferences and journals, including IEEE INFOCOMM, GLOBECOM, and TMC.

RECO : An Open-Source Reconfigurable Core Network

Prof. Jyh Cheng Chen ( National Chiao Tung University, Taiwan )

Abstract

It is envisioned in the future that not only smartphones will connect to cellular networks, but also all kinds of different wearable devices, sensors, vehicles, etc. However, since the characteristics of different devices differ largely, people argue that future 5G communication systems should be designed to elastically accommodate these different scenarios. We propose a reconfigurable core network to achieve the NGMN vision of slicing the mobile networks to suit for different types of users. We also built a prototype based on openair-cn to demonstrate the architecture we proposed. The source code is open to anyone freely.

Bio.

Jyh-Cheng Chen has been a Faculty Member with National Chiao Tung University (NCTU), Hsinchu, Taiwan since 2010. Prior to that, he was with Bellcore/Telcordia Technologies in New Jersey, USA, and National Tsing Hua University (NTHU), Hsinchu, Taiwan. He is also now serving as the Convener, Computer Science Program, Ministry of Science and Technology, Taiwan. Dr. Chen received numerous awards, including the Outstanding Teaching Awards from both NCTU and NTHU, the Outstanding Research Award from the Ministry of Science and Technology, the Outstanding I. T. Elite Award, Taiwan, the K. T. Li Breakthrough Award from the Institute of Information and Computing Machinery, and the Telcordia CEO Award. He is a Fellow of the IEEE and a Distinguished Member of the ACM. He was a member of the Fellows Evaluation Committee, IEEE Computer Society.

Towards a distributed SDN controller - ONOS

Dr. Jian Li ( POSTECH, Korea )

Abstract

In this talk, a summary of Open Networking Operating System (ONOS) will be given. ONOS is an open source distributed SDN controller project, started in 2013 at Open Networking Laboratory (ON.LAB), open sourced in 2014, and hosted by Linux Foundation. The objective of this project is to deliver a network OS to fulfill service providers' stringent requirements. ONOS aims to provide scalability, high availability, high performance and abstracted APIs to allow ONOS application developer easily develop their application and service that control the traffic in the network. Today the platform is based on a solid architecture written in Java, and has quickly matured to be feature rich and production ready. This talk will have a technical focus discussing the different aspects of the ONOS architecture, network state distribution, northbound APIs, southbound protocols integration and applications.

Bio.

Jian Li is a member of Ambassador Steering Team of Open Networking Foundation (ONF), mainly in charge of governing ONF Ambassador program. Jian Li is also currently serving as a chair of ONOS/CORD Working Group under SDN/NFV forum in Korea. He participates several ONOS brigade’s activities including performance and security brigade, SDN teaching brigade and localization brigade; and leads gRPC northbound interface brigade. Jian Li is a SDN/NFV enthusiast, mainly focuses on promoting open source SDN/NFV project (e.g., ONO and CORD) to Korea open networking academia and industry. He received his Ph.D degree from POSTECH in 2016. His research is in the area of Software-Defined Networks (SDN), mainly focuses on control plane management and Locator/Identifier Separation Protocol (LISP) design and implementation. Jian is an active ONOS developer, contributed code in various areas including Control Plane Manager (CPMan), northbound interfaces (e.g., REST, gRPC), and southbound interfaces (e.g., OpenFlow, LISP).
 

Fog/Edge computing Platform : Enabling Low-Latency Application in Next Generation Network

Dr. Yuan-Yao Shih ( Academia Sinica, Taiwan )

Abstract

Recently, with the ubiquitously connected smart devices, the Internet of Things (IoT) has received tremendous attentions and is considered as a promising architecture for many applications. With the diversity of the IoT applications, such as wearable computing, smart metering, smart home/city, vehicles and health monitoring, a large amount of dense, distributed, and mostly mobile IoT devices are expected for deployment shortly. In addition, many applications (such as augmented/virtual reality and vehicle automation) are demanding in terms of high bandwidth and low latency. These applications need intensive computations to accomplish object tracking, content analytics and intelligent decision for better accuracy, performance and user experiences. Current networking infrastructure, including radio access and backhaul, encounter difficulties in dealing with the increasing IoT traffics; thus, to fulfill the service requirements of those IoT applications, cloud computing is considered as a promising architecture, which can provide elastic resources to applications on the resource-limited IoT devices. However, many challenges remain unsolved, such as mobility support, location-awareness and ultra-low latency requirements due to possible long network delay in traversing the time-sensitive data traffics through the Internet backbone.
A new paradigm, called fog/edge computing, is emerging. It is an architecture by extending cloud computing to the edge of the network. Fog/edge computing has the potential to fulfill the ultra-low latency requirements for new rising machine-type communication (MTC) applications (such as tactile Internet, mobile augmented reality, and vehicle automation) by joint powerful computing of multiple fog/edge nodes and near-range communications at the edge. Since many emerging IoT applications require ultra-low latency, in addition to computing latency, communication latency also cannot be neglected. The location of fog/edge nodes and users be considered as a major factor when deciding which nodes supply resource to which users. The success of the fog/edge computing platform relies on the efficiency and sustainability of the platform. This talk will discuss how the fog/edge computing platform can enable ultra-low latency applications for next generation networks and the challenges to realize a successful fog/edge computing platform.

Bio.

Yuan-Yao Shih is currently a Postdoctoral Research Fellow at Research Center for Information Technology Innovation, Academia Sinica. He received his B.S. degree in Computer Science from National Tsing Hua University, Taiwan, in 2008, M.S. and Ph.D. degrees in Computer Science and Information Engineering from National Taiwan University, Taiwan, in 2010 and 2015 respectively. His Ph.D. thesis was awarded IICM Best Ph.D. Dissertation Award. He was a visiting student in the Department of Electrical Engineering at the Princeton University in 2015 and a visiting scholar in the Department of Computer Science at University of Minnesota from 2016 to 2017.

Cooperative safety with V2X for automated driving systems and some ADAS applications : use cases and

technologies

Dr. Si Bok Yu ( Korea Automotive Technology Institute (KATECH), Korea )

Abstract

In this presentation, the current development status of vehicle automation systems in Korea is shortly summarized. The system use cases and test case for C-FVCWS (Cooperative-Forward Vehicle Collision Warning Systems) is discussed. C-FVCWS is the V2V based system which provides the warning service for drivers when there is a risk of collision with the forward vehicle on the same path with the ego vehicle. The presentation also includes the development of the system standard for C-PADS(Cooperative - Partially Automated In-Lane Driving Systems) and the test cases for C-PADS. The C-PADS is the system which the longitudinal control of the automated vehicle is activated by the wireless communication from the road infrastructure. The development project for C-FVCWS was finished in 2015, and the C-PADS is now in the 3rd year development phase.

Bio.

Si-bok Yu, Ph.D., is the Director of Vehicle Autonomous Technology R&D Center in Korea Automotive Technology Institute(KATECH). He received the M.S. and Ph.D. from Aerospace Engineering department at Old Dominion University in Virginia.
His is currently focusing on the cooperative automated vehicle research. The research includes the longitudinal control of the automated vehicle based on the wireless communication with the road infrastructure. 
He has been working in KATECH since 2003. In 2015, he received the Ministry Award from MSIP (Ministry of Science, ICT and Future Planning). He has been working in ISO TC204 WG14 ‘Vehicle Active Control & Warning Systems’ since 2006, and he is now the Korean representative of the WG. He was the project leader for ISO 11067 ‘CSWS (Curve Speed Warning Systems), and he is now the project leader for ISO/NP 20901 ‘EEBL (Emergency Electronic Brake Light Systems).
 

App-Specific Edge Computing and In-Network Deep Learning

Prof. Akihiro Nakao ( University of Tokyo, Japan )

Abstract

Network softwarization and network slicing are considered essential concepts for building 5G mobile networks and beyond to flexibly accommodating applications and services with a wide spectrum of diverse requirements. Mobile/Multi-access Edge Computing is also emerging technology to deal with ultra-reliable and low-latency applications. In this presentation, we introduce our research on application specific network slicing enabling edge computing for a target application utilizing flexible network infrastructure brought by network softwarization. We apply deep machine learning within networks to identify applications from given flows of traffic to create slices per application and to enable execution environments for edge computing per application.

Bio.

Akihiro Nakao received his BS in Physics and ME in Information Engineering from the University of Tokyo. He worked at IBM Yamato Laboratory, Tokyo Research Laboratory, and IBM Texas Austin. He received his MS and PhD in Computer Science from Princeton University. Since 2005, he has been an Associate Professor and is now a Professor in Applied Computer Science at the Interfaculty Initiative in Information (III)  Studies, Graduate School of Interdisciplinary Information Studies, University of Tokyo. He has been appointed as Chairperson of Department in III. He has also been appointed Chairman of the 5G Mobile Network Promotion Forum (5GMF) Network Architecture Committee by Japanese government.

Wireless Information-Centric Networking with Edge Computing for Vehicular Applications

Prof. Byung-Seo Kim ( Hongik University, Korea )

Abstract

Future connected vehicular systems will be able to obtain massive driving environment data in real time and by analyzing such a massive information, vehicle itself will drive to a destination with avoiding upcoming hazards. To realize the system, the high computation capability, the low latency communication, and the more flexible distributed communication model are required. In regard to the computing capability, Cloud computing might be solution since it offers storage, computation and software services on demand in the Cloud. However, it might not meet the required latency because it locates way far from the vehicular system. Therefore, instead of Cloud computing, to provide both of high computing capability and low latency, Mobile Edge Computing (MEC) or Fog Computing might be a solution since this paradigm offers computation, data, and application services in close proximity of end users. Regarding the more flexible distributed communication model, Information-centric networking (ICN) might be an alternative since it provides non-host centralized network, non-IP-based data publish/subscribe mechanism, broadcasting-based transmissions, etc. In this talk, ICN approach with Mobile Edge computing (MEC) for vehicular communications are discussed. This talk will cover the fundamentals of ICN and MEC/Fog computing, use cases of vehicular applications over ICN with Edge Computing, issues for ICN over MEC, and wireless networks adopting ICN/MEC concepts.

Bio.

Byung-Seo Kim is an Associate Professor at the Department of Computer and Information Communication Engineering in Hongik University-Sejong Campus. He received his B.S. degree in Electrical Engineering from In-Ha University, Korea in 1998 and his M.S. and Ph.D. degrees in Electrical and Computer Engineering from the University of Florida in 2001 and 2004, respectively. From January 2005 to August 2007, he worked for Motorola Inc., Illinois and his researches in Motorola Inc. were about wireless broadband mission critical communications. He served as Guest Editors of special issues of International Journal of Distributed Sensor Networks, IEEE Access, and Journal of the Institute of Electrics and Information Engineers, and served as General Chair and TPC members for international conferences. He is IEEE Senior Member. His research interests include the design and development of efficient wireless/wired networks including link-adaptable/cross-layer-based protocols, wireless CCNs/NDNs, Mobile Edge Computing, physical layer for broadband PLC, and resource allocation algorithms. His works have appeared in around 141 publications and 22 patents.