Network Evolution in Coming 10 Years: What's the Future of

Download Report

Transcript Network Evolution in Coming 10 Years: What's the Future of

Network Evolution in
Coming 10 Years: What's
the Future of Network?
~Toward Information-Centric Networking~
Byeong-Gi KIM
Park Laboratory, GITS, Waseda University
October 2nd , 2013
Outline

Background: Information-Centric Networking

Introduction: Content-Centric Networking

Research Challenges

Research in Progress

Conclusion
Background

Information-Centric Networking (ICN)
• “Its founding principle is that a communication
network should allow a user to focus on the
content he or she needs, rather than having to
reference a specific physical information where that
content is to be retrieved from”
Request
Reply
User
Internet
Introduction

Content-Centric Networking (CCN)
CCN moves the universal component of the
network stack from IP packets to named con
tent chunks
Introduction (cont.)

Content-Centric Networking (cont.)
• 2 packet types
• Interest: used for requesting a specific content
• Data: sent back to the corresponding consumer(s) in
response to the Interest(s)
CCN packet types
Introduction (cont.)

Content-Centric Networking (cont.)
• 3 data structures
• Forwarding Information Base (FIB): used for forwarding
the matching Interests to the corresponding publisher(s)
• Content Store (CS): store the incoming content
• Pending Interest Table (PIT): record the incoming Interest
until the corresponding content is received
Introduction (cont.)

Content-Centric Networking (cont.)
Forwarding Information Base (FIB)
Name
Prefix
Outgoing
Interface
waseda.jp/
1
Data is stored in Content
Store (CS)
Interest
waseda.jp/sample.jpg
0
1
Data
Pending Interest Table (PIT)
Content
Name
Incoming
Interface
waseda.jp/sa
mple.jpg
0
Research Challenges

Content-oriented routing
• Current routing protocols are based on location
•

information, i.e., IP addresses.
Metrics that can reflect content itself, e.g., content
popularity, should be used for routing in CCN
Improving efficiency of cache usage
• Not only considering cache replacement algorithm
but also distributed caching algorithm need to be
researched.
Research Challenges (cont.)

Effective multi-face forwarding
• CCN supports multi-face packet forwarding, but there
has been few research how to use the multi-face
forwarding function effectively

Support current applications
• Currently, real-time streaming and content sharing
•
applications have been mainly researched.
The other applications further need to be investigated
and researched.
Research in Progress

Data Gathering
• 1 Interest-to-1 Data exchange communication in CCN
•
would have a bad influence in aspect of scalability in
large scale networks, i.e., sensor networks.
We aim at reducing the overhead such as control
traffic, energy consumption.
Research in Progress (cont.)

Push-oriented application support
• CCN architecture supports pull-oriented
•
communication, but it would cause large control
overhead while supporting push-oriented applications,
e.g., emergency announcement.
We are designing a protocol to support the pushoriented communication over CCN.
Conclusion

ICN is an alternative approach to the architecture of computer
networks, which aims at supporting current Interest users’ demands;
users care about data itself instead of where the data is.

CCN architecture provides information-centric principles with
bringing several strong points; native asynchronous/synchronous
multicast support, cache usage, security, etc.

There are many challenge issues;
•
•
•
•
Content-oriented routing
Improving efficiency of cache usage
Effective multi-face forwarding
Support current applications, e.g., real-time streaming.
Thank you!