IEEE NetSoft 2016

Kohei Shiomoto


Speaker: Kohei Shiomoto, NTT Labs, Japan

Title: NetroSphere and NetworkAnalytics – Innovation of network architecture and operation beyond SDN/NFV evolution


As a number of diversified network equipments are deployed in the carrier networks, we need to overcome the limit of development, technology, and operation by transforming from vertically-integrated silo-style development to horizontal collaborative open innovation. Software-Defined Networking (SDN) and Network Function Virtualization (NFV) is expected to create horizontal collaborative open innovation by enabling  softwarized functions running over commodity server and switch-router hardware. We are pursuing component-level disaggregation beyond SDN/NFV to create healthy sustainable ecosystem where new players can join without any barriers. NetroSphere is a paradigm where we deepen disaggregation into component-level and assemble components to provide agile and adaptive services to service provider and to reduce cost by commoditization. In order to reduce cost, support service creation, and improve resiliency and agility, NetroSphere introduces (1) Separation between optics and electronics, (2) Separation of functions from equipment, and (3) Separation of resources from equipment.

NetworkAnalytics is a key enabler of NetsroSphere concept, where we understand what’s going on network by analyzing diversified kinds of data (traffic measurement, QoE, configuration, error, SNS, etc.) collected from inside and outside of networks and to provide value to two-sided: end-users and service providers after we extract intelligence from analyzed data. However, today’s circumstances surrounding telecom carriers is getting complicated. Telecom carriers face two-sided situation where they have to provide value to both end-users and service providers. Service providers generate traffic source and have strong influence traffic demand matrix while end users enjoy fixed-mobile diversified services (web browsing, video streaming, etc.) making traffic prediction hard. Carrier network is growing and getting complicated with a number of diversified network equipments provided by multi-vendors. To cope with the complicated situation, we are pursuing inter-disciplinary research framework by combining research fields such as machine learning theory, control-theory, information theory in addition to optimization theory, graph theory, statistical analysis, queuing theory, etc. Traffic data, syslog data, and Twitter data is analyzed for various network operations: traffic-engineering, trouble-shooting, QoE-sentiment analysis. We call this inter-disciplinary as “NetworkScience”. where different field of technologies are employed in addition to existing network technologies. Machine-learning, information theory, control theory, etc. are introduced.

This keynote provided an overview of R&D activities on NetroSphere and NetworkAnalytics conducted in NTT R&D labs.


Kohei Shiomoto is Senior Manager of Communication & Traffic Service Quality Project, NTT Network Technology Laboratories, NTT, Tokyo, Japan.  He joined the Nippon Telegraph and Telephone Corporation (NTT), Tokyo, Japan in April 1989. He has been engaged in R&D of high-speed networking including ATM, IP, (G)MPLS, and IP+Optical networking in NTT labs. From August 1996 to September 1997 he was a visiting scholar at Washington University in St. Louis, MO, USA. Since July 2012, he has been leading Communication & Traffic Service Quality Project of NTT Network Technology Laboratories, NTT, Tokyo, Japan. He received the B.E., M.E., and Ph.D degrees in information and computer sciences from Osaka University, Osaka in 1987 1989, and 1998, respectively. He is a Fellow of IEICE, a Senior Member of IEEE, and a member of ACM.