grids at the high energy frontier - university of...

21 February 2005 John Huth, Harvard UniversityAnnual AAAS Meeting 05

Grids at the High Energy Frontier

John HuthHarvard University


The High Energy Frontier• High energy = short distance (10-19 cm)• Outstanding questions in particle physics:

– Origins of mass, unification of fundamental forces (Higgs particle?, other?)

– Quantum nature of gravity (extra dimensions?)– Composition of the universe (dark matter, energy)

• Large Hadron Collider– Collaborations of unprecedented size

• A vanguard in the evolution of science– Pushes the limits of scientific computing

• Use of grids to enable democratic access to data and computing resources


Periodic Table of Fundamental Particles

Families reflectincreasing mass and a theoreticalorganizationu, d, e are “normalmatter”.Because of the chargequarks, electrons, muons, and tau’sparticipate in EM

-1

+2/3

-1/3

0

Mass


What holds them together?

• Fundamental forces– Gravity– Electromagnetism– The Strong force (holds the nucleus together)– The Weak force (regulates the burning of

hydrogen into heavier elements in stars)


Unified Theories

At very high energies all interactions merge to a single strength.


We are on the threshold of a fundamental energy scale

• The structure of the theory only works if the “symmetry breaking” – or generation of mass is apparent at energy scales we are about to probe

• The Large Hadron Collider at CERN– Expected to begin collisions of protons on

protons in 2007


The Large Hadron Collider


Large Hadron Collider at CERN

pp collider = 14 TeVs


15 meter magnet


The ATLAS Detector


The CMS Detector

MUON BARREL

CALORIMETERS

Silicon MicrostripsPixels

ECALScintillating

PbWO4 crystals

SUPERCONDUCTINGCOIL

IRON YOKE

TRACKER

MUONENDCAPS

Total weight : 12,500 tOverall diameter : 15 mOverall length : 21.6 mMagnetic field : 4 Tesla

HCALPlastic scintillator/brasssandwich


ATLAS cavern yesterday


Challenges at the LHC

Petabytes/year of data logged

2000 + Collaborators

37 Countries

140 Institutions (Universities, National Laboratories)

CPU intensive

Global distribution of data

Test with « Data Challenges »


CPU v. Collab.

10

10 0

1, 0 0 0

10 , 0 0 0

10 0 , 0 0 0

0 5 0 0 1 0 0 0 1 5 0 0 2 0 0 0 2 5 0 0

C o l l a b o r a t i o n S i z e

C P U CPU v. Collab.

Earth Simulator

Atmospheric Chemistry Group

LHC Exp.

Astronomy

Grav. Wave

Nuclear Exp.

Current accelerator Exp.

CPU vs. Collaboration Size


Problem of data and resource sharing

• Large scientific collaborations need to share data globally– Share computing resources– Have applications run seamlessly in an

heterogeneous environment– Guarantee reproducibility of analyses– Intelligent scheduling across a wide-area-network

• Grids are the solution


Image courtesy Harvey Newman, Caltech

Grids for High Energy Physics

Tier2 Centre ~1 TIPS

Online System

Offline Processor Farm

~20 TIPS

CERN Computer Centre

FermiLab ~4 TIPSFrance Regional Centre

Italy Regional Centre

Germany Regional Centre

InstituteInstituteInstituteInstitute ~0.25TIPS

Physicist workstations

~100 MBytes/sec

~100 MBytes/sec

~622 Mbits/sec

~1 MBytes/sec

There is a “bunch crossing” every 25 nsecs.There are 100 “triggers” per secondEach triggered event is ~1 MByte in size

Physicists work on analysis “channels”.Each institute will have ~10 physicists working on one or more channels; data for these channels should be cached by the institute server

Physics data cache

~PBytes/sec

~622 Mbits/sec




Caltech ~1 TIPS

~622 Mbits/sec

Tier 0Tier 0

Tier 1Tier 1

Tier 2Tier 2

Tier 4Tier 4

1 TIPS is approximately 25,000 SpecInt95 equivalents


Grid3

virtual data researchend-to-end HENPapplications

CERN LHC: US ATLAStestbeds & data challenges

CERN LHC: USCMStestbeds & data challenges

virtual data grid laboratory


Service monitoring

Grid3 – a snapshot of sites

Sep 04•30 sites, multi-VO•shared resources•~3000 CPUs (shared)


Shared infrastructure

cms dc04

atlasdc2

Usa

ge: C

PU

s


Three active grids: Nordu-grid, Grid3 and LCG

-20000

0

20000

40000

60000

80000

100000

120000

140000

4062

340

626

4062

940

702

4070

540

708

4071

140

714

4071

740

720

4072

340

726

4072

940

801

4080

440

807

4081

040

813

4081

640

819

4082

240

825

4082

840

831

4090

340

906

4090

940

912

4091

540

918

Days

LCGNorduGridGrid3Total

# V

alid

ated

Job

s

total

Day


Scaling and the Future

• Need to reduce amount of human intervention• Establishment of an economic model for the grid

– What are the real prices of services?– Get beyond “good-will” stage

• Open Science Grid– Next step beyond Grid3

• Security• Data storage and access• Quality of service

• Interoperability among grids– Standards


Summary

• Challenges of high energy frontier– Scale of experiments– Democracy of access to data and resources

• Grid solution well matched to the needs of the collaborations– Many applications are now using this infrastructure –

biology, earth sciences etc• Future developments

– Scaling, robustness, interoperability, standards

grids at the high energy frontier - university of...

Documents