10 years ago

• Three major issues:

• Very hierarchical IP networks– Limited ability to innovate or support high end users

• Supporting high science users on campus– Firewalls, misconfigs, limited bandwidth

• Heterogeneous independent customer owned fiber networks– But how do you build end to end solutions from independent

federated networks?

Today’s (2001) hierarchical IP network

UniversityRegional

National or Pan-Nationl IP Network

Other national networks

NREN A NREN B NREN C NREN D

Tomorrow’s peer to peer federated optical networks

University Regional

Server

World World

National DWDM Open Lightpath Exchanges

NREN A NREN BNREN C NREN D

ChildLightpaths

Child Lightpaths

Thanks to GLIF and LHCONE

• Our vision for optical lightpath exchanges (GOLEs) is now becoming a reality

• LHCONE will drive this new network reality

• Hierarchical IP R&E networks will not disappear – but will be complemented by direct optical connections through GLIF and GOLEs especially for eScience

Carleton University

School of Architecture

PhysicsDepartment

Engineering Telecom

Firewall

Main campusNetwork

BorderRouter

University

Internet

Eucalyptus*Design Network

GlobalPhysicsNetwork

ResearchTestbed

NREN

Campus CWDM

10G

1G

1G

1G

10G

*Will connect to UCLA Architecture and Optiputer

University of British Columbia

Research Hospital

TRIUMF

Engineering Telecom

Firewall

Main campusNetwork

BorderRouter

University

Internet

Health Network

GlobalPhysicsNetwork

3D HDTV toMcGill

BCnet

CANARIE DWDMSwitch at UBC

5G

1G

1G

1G

3G

1GTier 1

Tier 2

CERN

ESnet Science DMZ

CRC-i2CAT-Inocybe-UofO

GENI and UCLP

Substrate Router

InstrumentGRIM WS

SubstrateSwitch

ParentLightpathWS

TimesliceWS

Child Lightpath WS(may run over IPEthernet, MPLS, etc

GMPLSDaemon WS

VirtualRouterWS

Wireless SensorNetwork

Creating Federated Networks

Domain C

Domain A Domain B

Multi-Domain APN

Partitioned NodePass Through Node

Multi-Domain APN

CRC-i2CAT-Inocybe-UofO

UCLP==OpenFlow?

GreenStar –Clouds and Virtualization

Distributed computing architectures, applications, grids, clouds, Web services, virtualization, dematerialization, remote instrumentation and sensors, etc.

Share infrastructure & maximize lower cost power by “following wind & sun” networks.

Develop benchmarking tools to earn CO2 offset dollars for university and ICT department

http://www.greenstarnetwork.com/

GreenStar Network• World’s first zero carbon network• Nodes in Ireland, USA Spain and

Belgium to be added shortly• http://www.greenstarnetwork.com/

What are the major future challenges and opportunities?

• Universities are under increasing financial pressure – Need to reduce costs across the board including fees to eScience

infrastructure– IT energy consumption 20-40%

• Increasing demand for more eScience research and applications

• Increasing demand for nation wide mobile seamless services especially for personal health applications

• Very little disaster planning, especially from climate change

My predictions -1

• R&E networks will need to partner with industry and campus IT to develop solutions for eScience and general IP to reduce staff workload and energy costs

• Campus IT don’t have technical resources or budget to deal with DNSESC, IPv6, eScience– Ottawa University recently contract with Bell Canada for

management of their 3500 Hotspots

• Federated optical network linking GOLEs will be dominant network architecture (perhaps based on OpenFLOW

My predictions -2• Climate change is a real and present threat

– We are on the cusp of major climatic disruption

• ICT has become the heavy industry of the information age– CO2 emissions from ICT exceed smelters, steel mills and cement plants– Continuing dramatic growth rate

• Energy efficiency is not enough– We don’t have enough time to reduce emissions through energy efficiency

• We have seen from Dallas Thorton’s presentation we can achieve 75% reduction in energy costs and 100% reduction in CO2 emissions by relocating data centers to colder climates– This represents tens if not hundreds of millions of dollars in annual savings– Energy efficiency will never achieve this

• ICT represents 20-40% of electrical energy consumption at a university– We need to find solutions so that money spent on energy consumption can be

redirected to support research and education