Advanced Networking for Researchers in Europe

Download Report

Transcript Advanced Networking for Researchers in Europe

ISOC 2002
Israel, 18th February 2002
Advance networking for
researchers in Europe
an IST perspective
Mário Campolargo
Deputy-Head of Unit, DG INFSO, EC
[email protected]
"The views expressed in this presentation are those of the author and do not necessarily reflect the views of the European Commission"
Contents
 Information Society in Europe
 eEurope and IST - Research Networks
 GÉANT and the global perspective
 Piloting GRIDs platforms
 Piloting the new protocol IPv6
 Conclusions
An ambitious strategic goal
for EU in the next decade
“…to become the most competitive
and dynamic knowledge-based economy
in the world…”
(Lisbon Summit 2000)
Specific measures:
eEurope Action Plan
Launched by the European Commission to
meet the new challenge
Stimulate the use of
the internet
Investing in people
and skills
A cheaper, faster,
secure internet
eEurope Action Plan:
The political context
Networks: Encourage and Innovate
REGULATORY
MEASURES




Unbundling of local loop
Spectrum allocation
Promote interoperability
….
TECHNOLOGY
INNOVATION




R&D
Research Networks
IPv6 testbeds
GRIDs ….
Research Networks:
Linking EU policies with R&D
eEUROP
E
R&D
IST Program
3,6 Beuro
POLICIE
S
Networks for research
Research on networks
Research Networks:
The complementary objectives
 Deployment of a world class network
matching the aggregated needs of
Europe’s Academic & Industrial Research
 Ensure synergies
 Use of advanced tesbeds for integration
& validation of next generation coms,
networking, applications & services
Research Networks:
The projects/initiatives
 GÉANT
Global connectivity -EUMEDIS, TEIN, ALIS
SERENATE strategic study
 GRIDS for scientific applications
IPv6 large scale testbeds
Optical routing, QoS, Digital Libraries, ...
Research Networks:
The funding from EU
 GÉANT
Global connectivity
MEuro
SERENATE strategic study
 GRIDS
MEuro
IPv6
Other
80 MEuro
20
1 MEuro
37
24 MEuro
8 MEuro
Research Networks:
The model
Tesbeds use
GÉANT infrastructure
GÉANT profits from
technological innovation
Scientific/application areas
GRIDs
IPv6
International
dimension
GÉANT network
GÉANT as the “flagship”
of networking activities in Europe
GÉANT - creating the pan-European
network infrastructure to explore
developments in telecommunications
technology.
 Network Services for National Research
and Education Networks (NRENs)
 Platform for testing - GRIDs, IPv6, ...
European
EU+ NRENs
GÉANT
National
NRENs
Campus
Universities
National funds
Other
funds
Global
Requiring new
levels of
cooperation
EU funds
GÉANT:
The model
GÉANT:
The European research backbone
 Pan-European coverage (and beyond)
 Interconnecting 32 NRENs (… Israel...)
 Linking more than 3000 Universities,
I.e. virtually all the researchers in
Europe in all disciplines
 9 international circuits operating at
10Gbps while 11 other run at 2,5Gbps
Total 200 MEuro over 4 years (80 Meuro
from EU)
GÉANT:
The connectivity at 10 Gbps
EE
LV
34
155
UK
155
LU
NL
155
155
BE
FR
SE
IE
LT
45
SE - PoP for Nordunet
155
DE
CZ
ES
PT
SK
622
CH
IT
AT
HU
622
34
HR
BG
622
622
GR
34
45
CY
155
SI
34
10 Gbps
2.5 Gbps
PL
IL
34
RO
GÉANT:
Access of NRENs to GÉANT
SE - PoP for Nordunet
NL
DE
CH
SE
HU
IT
FR
GR
CZ
BE
AT
UK
PT
GEANT
ES
SI
PL
IE
HR
LU
RO
LV
BG
CY
LT
IL
SK
EE
GÉANT:
The most advanced research backbone
NRENs* Access Capacity to the GEANT Backbone
Gigabits per sec
(June and December 2001)
5
4
3
2
1
0
622Mbps
DK FIN S
B
D
E
F
I
NL UK A
EL IRL L
P
Abilene
GÉANT
Trunk Capacity
35 GB/s
120 GB/s
No of Main
Access Points
36
27
No of Core
Nodes
13
12
Accessible
Institutions
200 aprox.
GÉANT +
NRENs
> 3000
GÉANT international dimension:
a world of opportunities
5Gbps
ALIS
EUMEDIS
2Mbps
TEIN
155Mbps
GÉANT:
Future challenges
 Consolidate outreaching
 Address the Terabits per second
 Enhance “inclusiveness” (educational
networks, libraries,...)
 Strengthen links with industry
 Anticipate “full” liberalization in
telecommunications
Disruptive
emerging
Technology
GÉANT:
One element of a strategy
Experimental
Advanced
IPv6,
Optical,...
GRIDS
GÉANT
All
research
disciplines
Very
demanding
communities
Research
on
networking
GRIDs:
A new dimension in networking
The GRID aims to create unprecedented
computing and information power by linking
individual machines over high-speed
networks using advanced middleware.
GRIDs:
Why do we need them… CERN...
 The Large Hadron Collider Project
CER
N
Storage –
Raw recording rate 0.1 – 1 Gbps
Accumulating at 5-8 PetaBytes/year
10 PetaBytes of disk
Processing –
200,000 of today’s fastest PCs
GRIDs:
Why do we need them… Industry...
 Automotive industry...
Mercedes Benz crash simulations
one run uses 100 h CPU
1.5 GB memory Storage
GRIDs:
A new dimension in networking
The concept of GRIDs addresses the
situation in which the bandwidth of the
network blurs the frontiers of the
computer…
GRIDs deploys technologies that give users
a seamless, secure access to a variety of
data and computing resources…
GRIDs integrate and adds value to
networking, computing, middleware,
security, knowledge engineering, ...
GRIDs:
Building virtual communities
GRIDs address the needs of virtual
communities and fosters the efficient
exploitation of the huge investments in
research infrastructures:
 High Energy Physics
 Earth & Space Observation
 Environment
 Bio-sciences and Health
 Industrial
design/simulation/visualization
 ASP
GRIDs:
The European efforts
EU supported GRIDs work complements
national Programmes:
UK - eScience Programme
 Italy - INFN Grid
 Netherlands - DutchGrid
 Germany - Unicore plus
 France - Grid funding approved
 Ireland - Grid-Ireland
 Poland, Czech Republic - seed
funding

GRIDs:
An integrated approach
Applications
EGSO
CROSSGRID
Middleware GRIP EUROGRID
& Tools
DAMIEN
GRIDSTART
GRIA
GRIDLAB
DATAGRID
DATATAG
Underlying
Infrastructures
Industry / business
Science
GRIDs:
The IST projects
 37Meuro of IST funding
 Strong consensus formation and
contribution to standards: GGF, IETF,
W3C, MPI, etc
 Most developments follow Open
Architecture approach.
GRIDs:
Examples of large testbeds
 EUROGRID, GRIP
Application requirements:
• 6 European
• Real-time, Resource
countries
brokerage, Portals,
• Globus/Unicore
Coupled applications
interface
GRIDs:
Examples of large testbeds
 DATAGRID, CROSSGRID
• 17 European
countries
• Collaboration
of more than
2000 scientists
Application requirements:
• Computing > 20 TFlops/s
• Downloads > 0.5PBytes
• Network speeds at 10
Gbps
GRIDs:
Examples of large testbeds
 DATATAG (cross-Atlantic testbed)
(2Gbps)
Links with US projects
(GriPhyN, PPDG, iVDGL,…)
GRIDs:
Creating a solid technology base
 Portals; Application interfaces
(including coupled/interactive
applications)
 Techniques for data search, mass
storage, data analysis
 Dynamic resource discovery and
scheduling; Resource brokerage.
 Connectivity services (e.g .advanced
network reservation, QoS, access
greater range of end-user facilities)
GRIDs:
Creating a solid technology base
 Security, including support for end-toend business processes.
 Management, with emphasis
accounting, performance.
on
 Monitoring, including advanced
visualization techniques.
 Interoperability (at various levels,
including Globus-Unicore interface,
Windows applications).
IPv6:
A new protocol for the new Internet
 Interconnecting a myriad of “personal”
devices ...
 Going mobile…
 … and “always-on”…
 Restoring the end-to-end model
 Consumer electronic products IP
enabled…
IPv6:
Business case
 IPv6 is designed to improve upon IPv4's
scalability, security, ease-ofconfiguration, and network
management…
 These issues are crucial to the
competitiveness and performance of all
types of network-dependent businesses.
IPv6:
IPv4 address space
 Risk of global IPv4 addresses becoming
critically scarce by 2005
 Uneven distribution of the address space
IPv4 addresses
allocation
ARIN:
74%
RIPE NCC: 17%
APNIC:
9%
ARIN
RIPE NCC
APNIC
IPv6:
A coordinated strategy
 Technological components
 Experimentation in large testbeds
 Integration of fixed and mobile
 Co-operation with IPv6 Forum
 Launching of IPv6 Task Force
 Communication to the Council and
Parliament
IPv6:
Why testbeds
 Considering the level of standardization,
the availability of technology, the scale
of current experimentation, the
emergence of applications and the need
to accelerate the uptake of this
technology…
Large scale testbeds, involving the right
set of actors, emerge as an ideal way to
structure, consolidate and integrate
European efforts on IPv6.
IPv6:
Major testbeds
To Japan
 6NET
 EURO6IX
To North
America
 Native IPv6 at:
34 Mbps, 155 Mbps, 2,5 Gbps
 Pan-European coverage
To Korea
FP6:
Research networks related work
 Establish a high capacity and high speed
communications network for all
researchers in Europe (GEANT) and
specific high performance Grids and testbeds (GRIDs).
 Create large-scale distributed systems &
platforms, including Grid-based systems
to solve complex problems in areas like
the environment, energy, health,
transport, industrial design.
300 MEuro
Conclusions
 EU is actively promoting Information
Society in Europe
 Networking for Research is a imp ortant
priority for EU policy
 With the support of the IST Programme
and in the context of the eEurope action
plan, a new generation backbone for
research has been launched
 In parallel, EU is investing in networking
technologies such as GRIDS to support
highly demanding user community
Thanks for your attention