1. LHC-Computing Grid LCG
„Eventually, users will be unaware they are using any computer
but the one on their desk, because it will have the capabilities
to reach out across the (inter-) national network and obtain
whatever computational resources are necessary”
(Larry Smarr and Charles Catlett, 1992)
Hans F Hoffmann, CERN
November 2002
SPC

2. High Energy Physics is leading the way in data intensive science
4 large detectors for the Large Hadron Collider (LHC)
CMS
ATLAS
LHCb
Storage –
Raw recording rate 0.1 – 1 GByte/sec
Accumulating data at PetaBytes/year ~ 20 million CDs each year
10 PetaBytes of disk
Processing –
150,000 of today’s fastest PCs
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

3. Grids Mobile Access Supercomputer, PC-Cluster G R I D M L E W A
Desktop
Data-storage, Sensors, Experiments
Hoffmann, Reinefeld, Putzer
Internet, networks
Visualising
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

4. DESY-Nov-02; H F Hoffmann/CERN
LCG
Building a Grid
Grid
The virtual LHC Computing Centre
Collaborating
Computer
Centres
Alice VO
CMS or Atlas or LHCb VOs
What are the essential points of the Grid?
the Grid hides the complexity of the structure from the user and from the application
enables a single distributed facility to be run as a service for several experiments, several sciences
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

5. Virtual Computing Centre
LCG
Virtual Computing Centre
The User ---
sees the image of a single cluster
does not need to know - where the data is
- where the processing capacity is
- how things are interconnected
- the details of the different hardware
and is not concerned by the conflicting policies of the equipment owners and managers
should be as simple to use as a shared service like LXBATCH at CERN
should be as reliable and predictable as LXBATCH
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

6. What is the LHC Computing Grid Project?
LCG
What is the LHC Computing Grid Project?
Mission of Phase 1 – prepare the computing environment for the analysis of LHC data
applications software - environment, tools, frameworks, common developments
deploy and coordinate a global grid service
encompassing the LHC Regional Centres
coordination of significant efforts in middleware support, with emphasis on robustness, monitoring, error recovery (resilience)
strategy and policy for resource allocation
authentication, authorisation, accounting, security
monitoring, modelling & simulation of Grid operations
tools for optimising distributed systems
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

7. DESY-Nov-02; H F Hoffmann/CERN
LCG
CERN will provide the data reconstruction & recording service (Tier 0) -- but only a small part of the analysis capacity
current planning for capacity at CERN + principal Regional Centres
2002: 650 KSI  <1% of capacity required in 2008
2005: 6,600 KSI2000  < 10% of 2008 capacity
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

8. Centres taking part in the LCG-1
around the world  around the clock
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

9. SC2 PEB Project structure POB Grid Deployment Architects Board Forum
LCG
Project structure
POB
SC2
PEB
requirements
Architects
Forum
Design decisions,
implementation strategy
for physics applications
Grid Deployment
Board
Coordination, standards,
management policies
for operating the
LCG Grid Service
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

10. DESY-Nov-02; H F Hoffmann/CERN
LCG
LCG: SC2 (+ PEB) Roles
SC2 – Software&Computing Committee
PEB – Project Execution Board
SC2 brings together the four experiments and Tier 1 Regional Centers
It identifies common solutions and sets requirements for the project
may use RTAGs – Requirements and Technical Assessment Groups
limited scope, two-month lifetime with intermediate report
one member per experiment + experts
PEB manages the implementation
organizing projects, work packages
coordinating between the Regional Centers
collaborating with Grid projects
organizing grid services
SC2 approves the work plan, monitors progress
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

11. SC2 Monitors Progress of the Project
LCG
SC2 Monitors Progress of the Project
Requirements for several key work packages were successfully defined
PEB has turned these into work plans:
Data Persistency, Software Support, Mass Storage
Other work plans are in preparation:
Grid use cases, Mathematical Libraries
Key requirements are finishing in October:
Detector Simulation
‘Blueprint for LHC experiments software architecture’
This will trigger important further activity in application developments
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

12. Project Execution Board
LCG
Project Execution Board
Two bodies set up to coordinate & take decisions:
Architects Forum
software architect from each experiment, the application area manager
makes common design decisions and agreements between experiments in the applications area
supported by a weekly applications area meeting open to all participants
Grid Deployment Board
representatives from the experiments and from each country with an active Regional Centre taking part in the LCG Grid Service
prepares the agreements, takes the decisions, defines the standards and policies that are needed to set up and manage the LCG Global Grid Service
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

13. LCG Recruitment status - snapshot
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

14. DESY-Nov-02; H F Hoffmann/CERN
LCG Materials at CERN
LCG
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

15. DESY-Nov-02; H F Hoffmann/CERN
LCG
LCG Level 1 Milestones
Hybrid Event Store available for general users
applications
Distributed production using grid services
Distributed end-user interactive analysis
Full Persistency Framework
2002
2005
2004
2003
Q1 Q2 Q3 Q4
grid
LHC Global Grid TDR
“50% prototype” (LCG-3) available
LCG-1 reliability and performance targets
First Global Grid Service (LCG-1) available
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

16. ATLAS DC1 Phase 1 : July-August 2002
LCG
ATLAS DC1 Phase 1 : July-August 2002
3200 CPU‘s
110 kSI95
71000 CPU days
39 Institutes in 18 Countries
Australia
Austria
Canada
CERN
Czech Republic
France
Germany
Israel
Italy
Japan
Nordic
Russia
Spain
Taiwan UK
USA
grid tools used at 11 sites
5*10*7 events generated
1*10*7 events simulated
3*10*7 single particles
30 Tbytes
files
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

17. ATLAS DC2 : October 2003 - March 2004
LCG
Use Geant4
Perform large scale physics analysis
Use LCG common software
Use widely Grid middleware
Further test of the computing model
~ same amount of data as for DC1
ATLAS DC3 : End Begin 2005
5 times more data than for DC2
ATLAS DC4 : End Begin 2006
2 times more data than for DC3
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

18. DESY-Nov-02; H F Hoffmann/CERN
LCG
6 million events
~20 sites
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

19. DESY-Nov-02; H F Hoffmann/CERN
LCG
State of play
We are still solving basic reliability & functionality problems
This is worrying as we still have a long way to go to get to a solid service
At end 2002, a solid service in mid-2003 looks (surprisingly) ambitious
HEP needs to limit divergence in developments.
Complexity adds cost
We have not yet addressed system level issues
How to manage and maintain the Grid as a system providing a high-quality reliable service.
Few tools and treatment in current developments of problem determination, error recovery, fault tolerance etc.
Some of the advanced functionality we will need is only being thought about now
Comprehensive data management, SLA’s, reservation schemes, interactive use.
Many many initiatives are underway and more coming.
How do we manage the complexity of all this ?
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

20. Establishing Priorities
LCG
Establishing Priorities
We need to create a basic infrastructure that works well.
LHC needs a systems architecture and high-quality middleware – reliable and fault tolerant.
Tools for systems administration.
Focus on mainline physics requirements and robust data handling.
Simple end-user tools that deal with the complexity.
Need to look at the overall picture of what we are trying to do and focus resources on key priority developments
KISS
We must simplify and make the simple things work well.
It is easy to expand scope, much harder to contract it !
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

21. DESY-Nov-02; H F Hoffmann/CERN
LCG
Summary on LCG
basic middleware partnership
science, computer scientists and software engineers
industry and scientific research
international collaboration
global grid infrastructure for science
built on a foundation of core nodes serving existing national and global collaborations
advanced middleware research programme
complementary projects
with a focus on the emerging requirements of the LHC data analysis community
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

22. What else is going on
The International Context with the examples Europe and US
(Not mentioned here: UK: e-science, DE: FZK, Nordu-Grid, )

23. HEP related Grid projects
GriPhyN
PPDG
iVDGL
Notare la complementarieta’ dei paesi di appartenenza dei partner di progetto
Through links between sister projects, there is the
potential for a truely global scientific applications grid
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

24. DESY-Nov-02; H F Hoffmann/CERN
Project start , duration 3 years, 21 partners
Deliverables: Middleware tried out with Particle Physics-, Earth Observation-, Biomedical Applications
1st review successfully passed,production quality middleware asked by Expts
Applications - Distributed Fabrics - "Transparent" Grid Middleware
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

25. The benefits of building Grids: an example from Astronomy
Crab Nebula viewed
At four different
wavelengths: X-ray,
optical, infrared, radio.
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

26. The benefits of building Grids: an example from Earth observation
From global weather information to
precision agriculture and emergency response
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

27. Managing Large Scale Data
Hierarchical Storage and Indexing
Highly Distributed Source
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

28. Berkeley workshop (12-13 Nov 2002)
EU perspectives on Research on Grids
Kyriakos Baxevanidis
European Commission, DG INFSO

29. EU level effort complements National efforts
Foster cohesion, interoperability, cross-fertilization of knowledge, economies of scale, critical mass
Increase value
Multiply impact EU
(IST-Programme: the “flagship”)
National

30. Overview of EU level RTD-policy on Grids
Committed to develop, deploy Grids
Strong support to synergetic and integrated approaches
Strong support to international co- operation (EU with the other regions)

31. Major Infrastructure deployments (on-going)
1. DATAGRID, CROSSGRID, DATATAG
Examples:
Infrastructure across
17 European States
Cross-Atlantic link
Application requirements:
Computing > 20 TFlops/s
Downloads > 0.5PBytes
Network speeds at 10Gbit/s
Collaborations of more
than 2000 scientists
(2.5 Gbit/s)
2. EUROGRID, GRIP
Infrastructure across
6 European States
Industrial design, simulations
(as well as scientific applications)
Globus - Unicore interface
Application requirements:
Real-time, Resource brokerage,
Portals, Coupled applications
Links to National Grid
infrastructures

32. Grids in FP6: some important priorities
From current prototype to production level systems (industrial quality); promote commercial uptake
Research on new concepts (Semantic Grids, Grids and the Web, Mobile Grids, Ambient Intelligence Spaces)
Strengthen middle-ware developments in Europe (Academia-Industry collaboration, skills)
Involve new User/Application communities
Funding targeted more to big scale efforts (tens of millions of EURO) - mobilize National, private funds
Close ties with National efforts (National Grid Centres?)
Strengthen International cooperation - standards

33. Synergetic work: at the core of the activity
GRIDLAB
GRIA
EGSO
DATATAG
CROSSGRID
DATAGRID
GRIP
EUROGRID
DAMIEN
G R I D S T A R T
AVO
Synergy with new Grid projects, EU National and International efforts, GGF

34. DESY-Nov-02; H F Hoffmann/CERN
EGEE :EU 6th Framework Programme (Enabling Grids for E-science and industry in Europe)
EU and EU member states major investment in Grid Technology
Several good prototype results
Next Step:
Leverage current and planned national programmes
work closely with relevant industrial Grid developers and NRNs
build on existing middeware and expertise
create a general European Grid production quality infrastructure
This can be achieved for a minimum of €100m/4 years on top of the national and regional initiatives
applications
EGEE
network
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

35. Blue Ribbon Panel on Cyberinfrastructure Presentation to MAGIC
Paul Messina
November 6, 2002 1

36. Cyberinfrastructure: the Middle Layer
Applications in science and engineering research and education
Cyberinfrastructure: hardware, software, personnel, services, institutions
Base-technology: computation, storage, communication

37. Some roles of cyberinfrastructure
Processing, storage, connectivity
Performance, sharing, integration, etc
Make it easy to develop and deploy new applications
Tools, services, application commonality
Interoperability and extensibility enables future collaboration across disciplines
Best practices, models, expertise
Greatest need is software and experienced people

38. From Prime Minister Tony Blair’s Speech to the Royal Society (23 May 2002)
What is particularly impressive is the way that scientists are now undaunted by important complex phenomena. Pulling together the massive power available from modern computers, the engineering capability to design and build enormously complex automated instruments to collect new data, with the weight of scientific understanding developed over the centuries, the frontiers of science have moved into a detailed understanding of complex phenomena ranging from the genome to our global climate. Predictive climate modelling covers the period to the end of this century and beyond, with our own Hadley Centre playing the leading role internationally.
The emerging field of e-science should transform this kind of work. It's significant that the UK is the first country to develop a national e-science Grid, which intends to make access to computing power, scientific data repositories and experimental facilities as easy as the Web makes access to information.
One of the pilot e-science projects is to develop a digital mammographic archive, together with an intelligent medical decision support system for breast cancer diagnosis and treatment. An individual hospital will not have supercomputing facilties, but through the Grid it could buy the time it needs. So the surgeon in the operating room will be able to pull up a high-resolution mammogram to identify exactly where the tumour can be found.

39. Futures: The Computing Continuum
Petabyte
Archives
Smart
Objects
National
Petascale
Systems
Ubiquitous
Sensor/actuator
Networks
Collaboratories
Responsive
Environments
Terabit
Networks
Laboratory
Terascale
Systems
Contextual
Awareness
Ubiquitous Infosphere
Building Up
Building Out
Science, Policy
and Education

40. Key Points About the Proposed Initiative
There is grass roots vision and demand from broad S&E research communities. Many needs will not be met by commercial world.
Scope is broad, systemic, strategic. A lot more than supercomputing. Extreme science - not flops. Potential to relax constraints of distance, time, and disciplinary boundaries. New methods: computation, visualization, collaboration, intelligent instruments, data mining, etc.
Opportunity to leverage significantly prior NSF and other government investments. Potential large opportunity cost for not acting soon.
The initiative is intrinsically international: cooperation and competition. Can’t assume US is in the lead.

41. Components of CI-enabled science & engineering
A broad, systemic, strategic conceptualization
High-performance computing
for modeling, simulation, data
processing/mining
Humans
Instruments for
observation and
characterization.
Individual &
Global
Connectivity
Group Interfaces
Physical World
& Visualization
Facilities for activation,
Collaboration
manipulation and
Services
construction
Knowledge management
institutions for collection building
and curation of data, information,
literature, digital objects

42. Coordination (synergy) Matrix
Research in technologies, systems, and applications
Development or acquisition
Operations in support of end users
Applications of information technology to science and engineering research
Cyberinfrastructure in support of applications
Core technologies incorporated into cyberinfrastructure

43. Bottom-line Recommendations
NSF leadership for the Nation of an INITIATIVE to revolutionize science and engineering research capitalizing on new computing and communications opportunities.
21st Century Cyberinfrastructure includes supercomputing massive storage, networking, software, collaboration, visualization, and human resources
Current centers (NCSA, SDSC, PSC) and other programs are a key resource for the INITIATIVE.
Budget estimate: incremental $1020 M/year (continuing).

44. Basic Middleware Partnership
LCG
Basic Middleware Partnership
System Design
Group
Modelling
Project management
Software Process
Standards
Integration,
test,
certification
Product
Delivery
Middleware I II
System
Tools
End User Tools
Design
Production
Development
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

45. Basic Middleware Partnership
LCG
Basic Middleware Partnership
industrial management, development process – s/w engineering focus
few development centres – clear division of responsibilities
including key technology owners
close co-operation between US, European, Asian(?) teams design and implementation
strong preference for a single project – one management one review process multiple funding sources
an international partnership of – computer science and software engineering science and industry
System Design
Group
Modelling
Project management
Software Process
Standards
Integration,
test,
certification
Product
Delivery
Middleware I
Middleware II
System Tools
End User Tools
Design
Production
Development
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

46. The LHC Grid – the foundation of a Global Science Grid
LCG
The LHC Grid – the foundation of a Global Science Grid
LHC has a real need and a reasonable scale and has mature global collaborations of scientists
establish the basic middleware for a global science grid
deploy a core grid infrastructure in Europe, America, Asia
learn how to provide a “production quality” service during LHC preparation (data challenges)
exploit the LHC core infrastructure as the foundation for a general science grid
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

47. Development of a Global Grid Infrastructure for Science
LCG
Development of a Global Grid Infrastructure for Science
middleware
research
application-specific
middleware
lhc applications
science
application
adaptation
application-specific
middleware
bio applications
science
application
adaptation
application-specific
middleware
other science
applications
science
application
adaptation
computer science
providing solutions for exploitation of grids by scientists
…..
science research
Hardening/Reworking
of basic
middleware
prototyped by
current projects
Advanced
Middleware
requirements defined
by current projects
engineering
core grid infrastructure
few core centres (including the LCG Tier 1s)
operate the information service, catalogues, ..
coordination, operations centre, ..
call centre, user support, training, ..
converge with other core infrastructures (DTF, UK e-science grid, ….)
other grid nodes for
physics,
biology, medicine, ….
infrastructure
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

48. Longer-term Requirements
LCG
Longer-term Requirements
emphasis so far has been on the basic infrastructure - as a solid base for constructing the LHC analysis environment
building on this foundation we need a programme of collaborative/complementary research projects on advanced middleware
advanced collaborative environments (dynamic workspaces for scientific analysis communities)
resource optimisation – computation, storage, network
data placement, clustering, migration strategies
grid enabled object management
autonomic management and operation of grid resources
target 2009 –
a computing environment for LHC analysis teams, efficiently and effortlessly exploiting global scientific computing resources,
the scientist is not aware that she is using a grid
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

49. CERN users to be connected by a computer Grid
637 70
4306 22
538 87 55 27 10
Europe: institutes, 4603 users
Elsewhere: 208 institutes, 1632 users
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN

50. Prototyping the Global Information Society
“Bright” World to offer easy, affordable participation to e-sciences, publications, results, e-education to all interested countries
25-Nov-2002
DESY-Nov-02; H F Hoffmann/CERN