Transcript HIIT-NRC Seminar

[email protected]
HIIT’s Future Internet Research Programme
Kimmo Raatikainen
Programme Director: Future Internet
[email protected]
FI Team
• Director: Prof. Kimmo Raatikainen
• Research co-ordinator: Oriana Riva
• Seniors:
– Adj. Prof. Patrik Floréen
– Adj. Prof. Andrei Gurtov
– Dr. Arto Karila
– Dr. Kristiina Karvonen
– Univ. Lect. Markku Kojo
– Prof. Jukka Manner
– Dr. Pekka Nikander
– Dr. Ken Rimey
– Adj. Prof. Sasu Tarkoma
– Prof. Antti Ylä-Jääski
The Vision of the Future
• User expectations:
– Future applications and platforms will be contextsensitive, adaptive, and personalized
– They need to be run, in a reasonable and secure
manner, on a variety of execution environments:
anywhere, anyhow, anytime, by anyone
• Required system properties:
– self-aware, distributable, reconfigurable, proactive,
collaborative, secure, trusted, privacy providing,
mobile, diversely accessible, extendable,
incrementally deployable, resource-aware, …
Research Challenges in Future Internet
• Research Challenges:
– Security-Trust-Privacy
– Mobile Always-on Connectivity
– Scalable Open Service Architectures
• Solutions are sought in distributed algorithms and
structures, middleware, and protocols.
Dual Approach to Future Internet
• Improving current Internet
– Protocol enhancements: HIP, TCP, DCCP, SIP, etc
– Overlay networks
– Secure push
• Starting from clean table
– Publish-subscribe paradigm
– Applying microeconomics and game theory
Current Projects
PSIRP: Publish-Subscribe
Internet Routing Paradigm
UbiLife
Future
Mobility
Middleware
AwissNet: Ad-hoc PAN &
WIreless Sensor SEcure NETwork
Trust for All
Web Services
In Ad Hoc and
Mobile Infra
Location Privacy and Authentication
In Massively Distributed Systems
Trustworthy Internet
InfraHIP
NordicHIP
Finland-ICSI
Center for Novel
Internet Architectures
M
E
R
C
O
N
E
A
B
I
I
N
H
O
N
E
T
S
Security-Trust-Privacy
Andrei Gurtov
Adj. Prof.
Research Challenges
• Secure mobility and multihoming
– Secure mapping between identifiers and locators
– Efficiency and scalability of DHT and DNS systems
• Eavesdropping/modification of network traffic
– Lack of IPsec deployment
• Denial-of-service attacks
• Tracking user identity and location
• Troubles of SSL/TLS certificate model
• Email SPAM
• Usability of security for a wide range of people
• Interoperating IPv4 and IPv6 applications and networks
Main Achievements
• Open-source implementation of Host Identity Protocol
(HIPL)
– Prevents eavesdropping and some DoS attacks
– Secure mobility/multihoming, middlebox friendly
– New IPsec mode integrated into standard Linux
kernel
– Chairing research group in IRTF, IETF drafts
• Scalability and performance results of DHT/rendezvous
systems using Planetlab testbed
• International collaboration network (Berkeley, EU)
– Article at Sigcomm’07 on Delegation-Oriented
Network Architecture
Future Directions
• Internal HIP deployment in the group’s computers
• Secure VoIP architecture/demo using P2PSIP model
• HIP on Platforms implementation (Symbian/Internet
tablet)
– IP security on lightweight devices
• Applying game theory to model economic aspects of
security
• Evaluating security and mobility in DHTs
• Distributed trust, discovery of trust paths in community
• Usability evaluations with user experiments and
interviews
Mobile Always-on Connectivity
Jukka Manner
Professor, PhD.
Research Challenges
• Decentralized IP host mobility in radio mesh networks
• Next generation network control signalling framework
• Congestion controlled datagrams (DCCP) for VoIP
• SIP-based services for mobile nodes
• Heterogeneous traffic in heterogeneous multi-access
environments
• Reducing signaling overhead in wireless networks
• Cross-layer design and inter-layer co-operation
Main Achievements
• Many IETF contributions
• World first high performance GIST implementation (C)
• A new RSS push service for the N800 tablet based on SIP
• Improvements to TCP congestion control for wireless and
mobile environments using cross-layer design
– Algorithms implemented in Linux TCP/IP stack
• Brought Linux TCP stack into compliance with IETF
specifications
• DCCP code into the Linux kernel (forthcoming)
Future Directions
• Continue work with
–
–
–
–
Next generation network signalling
DCCP and SIP
Advanced end host mobility in mesh networks
Transport protocol improvements in multi-access
environments and vertical handoffs
• New congestion control algorithms and transport
protocols for heterogeneous traffic
• Future Internet architectures
– Implications of publish/subscribe paradigm
– A topic of the Finland-ICSI Center
Scalable Open Service Architectures
Sasu Tarkoma
Adj. Prof.
Research Challenges
• Evolution towards content-centric networking
– Shift towards multicast/anycast messaging
– Publish/subscribe for efficient asynchronous
communication
• Service delivery scalability to millions of customers
– Communication and maintenance cost of services
like YouTube are prohibitive
– Peer-to-peer technologies can help reduce network
and hosting costs and improve scalability
• Architectures for easy and secure service access
– Open, decentralized identity management (OpenID)
– Seamless Sign-On and easy identity bootstrapping
Main Achievements
• Middleware systems
– Wireless SOAP, content-based router, XMLaware data synchronizer, mobile desktop search
– Fuego and PDIS/S4All software used in other
projects, also outside HIIT
• Contributions to W3C Efficient XML Interchange
Working Group, WWRF Service Architecture
• Public demonstrations and presentations
Future Directions
• Publish/Subscribe Internet Routing Paradigm
– Develop new protocol stack for pub/sub data-centric
routing and forwarding in Internet scale
– EU FP7 STREP project PSIRP coordinated by HIIT
• Data structures and algorithms for routing in overlay
networks
• Peer-to-peer video-on-demand and video streaming
– BitTorrent for browsers
• Mobile AJAX: Asynchronous mobile applications
• Social networks and incentives for peer-to-peer
– Content-centric networking for social collaborative peerto-peer web