Transcript 20070716-boyd-summerhill

A Technology Vision for the Future
Rick Summerhill, Chief Technology Officer,
Eric Boyd, Deputy Technology Officer,
Internet2
Joint Techs Meeting
16 July 2007
FERMI Lab, IL
We Need a Vision for the Future
• Over the past 3-4 years, our community has been
focused on politics and costs - in particular tactical
issues that beg for a long term direction for technology
in the R&E community!
• In the past, our community has provided strong
leadership, but, of late, has focused on existing
services.
• We need to do more!
• And recalling history of previous projects and the
evolution of IP is important to understand
• It is time for our community to provide strong
leadership and direction
2
Important Ideas
• There are two fundamental ideas to
understand
• The most important role we can play is innovation you are the RONS, universities, and corporations
interested in doing new things !
• A vision for the future - in 5 years, networking will
have changed drastically, and in 10 years, it will be
unrecognizable!
• We need to focus on the future!
3
Community Assets and Infrastructure
• Our community now has available layer 0
through layer 3 services - and at the national,
regional, and campus levels
• It has obtained incredible resources and
infrastructure making new services and
capabilities available to the community
• At the National Level, for example, Internet2
now has full control of waves, both SONET
and Ethernet capabilities, and an IP network.
4
5
6
7
8
9
Need for Innovation and Direction!
• We need to create innovative ways to utilize the
infrastructure and provide new services! We need a
carefully thought out technology direction!
• From a community engineering point of view
• From a network research point of view
• Moreover, it’s not just the network, but a systems approach to
the complete Cyberinfrastructure
• Note that others in the international R&E community,
including partnerships with networks like ESnet, are
working very hard at this!
• Moreover, corporate enterprises and standard’s bodies
are closely watching what we do in this space.
10
Systems Approach
• There are many components to this approach
• Creating new network paradigms
• With the appropriate middleware components that
can support those network paradigms as well as
applications at higher layers in the stack
• With the appropriate performance monitoring and
diagnostics systems, again at all layers of the stack
• With the appropriate security mechanisms at all
layers
• A Systems approach to CyberInfrastructure
11
Technology Vision
• Over the next several months, create a
technology vision
• Provide direction for the future
• Will be a living document that will insure innovative
use of future developments
• Make clear what we do, and also what we don’t do
• Make clear which elements of the community do
which pieces, and how partnerships collaborate!
• Essential in this dialog is community input and
advice in creating that vision!
• We look to the councils and committees like the
NTAC for input!
12
Are you ready for LHC?
13
Local Infrastructure
US Tier 4
(1500 US
scientists)
Scientists
Analyze Data
US Tier 3
(68 orgs)
Internet2/Connectors
Internet2/Connectors
Scientists
Request Data
Atlas (6-7)
CMS (7)
US Tier 2
(15 orgs)
Provides
Data to
Tier 3
GEANT-ESNet-Internet2
Tier 1
(12 orgs)
FNAL
BNL
Shared Data Storage
and Reduction
LHCOPN
Tier 0
14
14
CERN
Raw Data
Peak Flow Network Requirements
Local Infrastructure
Tier 1 or 2 to Tier 3: Estimate: Requires 1.6 Gbps per transfer (2
TB's in 3 hours)
Internet2/Connectors
Internet2/Connectors
Tier 1 to Tier 2: Requires 10-20 Gbps
GEANT-ESNet-Internet2
Tier 0 to Tier1: Requires 10-40 Gbps
LHCOPN
CERN
15
15
What are the Implications for Normal
Network Operations from T2 to T3?
Example: 13 people (3 Professors and 10 Graduate Students)
require ten 3-hour timeslots a month to receive 8 Gigabit data
flows.
10 Gig
4 Gig
16
16
CMS T2 Traffic at UNL
17
Internet2 Connectors
CalREN-2 South
NYSERNet
3ROX
Great Plains Network
Indiana GigaPoP
MAGPI
MREN
Internet2
NoX
Merit
OARnet
ESnet
Oregon
GigaPoP
LONI
SoX
18
18
OmniPoP
Pacific Northwest
GigaPoP
Nine Universities Connect through
CalREN-2 South
UC Irvine (T3)
UC Santa Cruz (T3)
UC Davis (T2)
UCLA (T3)
University of
Arizona (T3)
UC Riverside (T3)
CENIC
UC San Diego(T3)
California Institute of Technology (T2)
UC Santa Barbara (T3)
19
19
The Network is the Backplane for the
Distributed CI Computer
Instrumentation
Control
Help
Desk
Researcher
Security
Viewing
Data
Generation
Program
Security
Collab
Tools
Control
Training
Education
And
Outreach
Publishing
Policy and
Funding
Management
Security and
Access
Access
Control
Human
Support
Funding
Agencies
Authentication
Resource
Providers
Campuses
3D
Imaging
Display
Tools
Authorization
Security
Security
Analysis
Simulation
Computation
Program
20
Input
Archive
Retrieval
Data Sets
Storage
Search
Data
Directories
Schema
Metadata
Ontologies
Data
.
Input
Network
Security
Display and
Visualization
The Network is the Backplane for the
Distributed CI Computer
Instrumentation
Control
Help
Desk
Researcher
Security
Viewing
Data
Generation
Program
Security
Collab
Tools
Control
Training
Education
And
Outreach
Publishing
Policy and
Funding
Management
Security and
Access
Access
Control
Human
Support
Funding
Agencies
Authentication
Resource
Providers
Campuses
3D
Imaging
Display
Tools
Authorization
Security
Security
Analysis
Simulation
Computation
Program
21
Input
Archive
Retrieval
Data Sets
Storage
Search
Data
Directories
Schema
Metadata
Ontologies
Data
.
Input
Network
Security
Display and
Visualization
Cyberinfrastructure Requirements
•
•
•
•
•
22
22
Data storage
Robust campus infrastructure
Security and Authorization
IT support for local and remote resources
Network Performance monitoring tools
Cyberinfrastructure View of the Network
CI Network
Internet2
23
Cyberinfrastructure Components
Applications
Bulk
Transport
2-Way
Interactive
Video
Real-Time
Communications
….
Phoebus
Middleware
….
….
Performance
Infrastructure / Tools
Network
24
….
Control
Plane
Network
Cyberinfrastructure
Applications call on Network Cyberinfrastructure
Questions?
[email protected]
25