Upload
openstack-foundation
View
573
Download
2
Tags:
Embed Size (px)
DESCRIPTION
Citation preview
Clouds in
High Energy Physics
Randall SobieInstitute of Particle Physics
University of Victoria
Randall Sobie IPP/UVictoria 2
High Energy Physics(Particle Physics)
The area of physics that studies the fundamental particles of nature and their
interactions.
SLAC Linear Accelerator
Accelerators
Sudbury Solar Neutrino Detector
Underground labs
Alpha Magnetic Spectrometer Space Station
Orbiting labs
Randall Sobie IPP/UVictoria 3
Is the Higgs boson the source of mass of our fundamental
particles?
Randall Sobie IPP/UVictoria 4
Why is the universe made of matter
and not equal amounts of matter/antimatter?
Randall Sobie IPP/UVictoria 5
What is origin of Dark Matter and Dark Energy?
TWe do not know the
composition of 95% of the universe
Temperature of the universeWMAP satellite
Randall Sobie IPP/UVictoria 6
Understanding our World
Randall Sobie IPP/UVictoria 7
Large Hadron Collider at CERN in Geneva
27 km in circumference100 m underground
Probes distances of 10-20 m
Randall Sobie IPP/UVictoria 8
Blue tubes contain the two beam pipes and magnets at 1.8 degrees Kelvin
Randall Sobie IPP/UVictoria 9
ATLAS Detector
44 m wide25 m diameter
Randall Sobie IPP/UVictoria 10
ATLAS detector during construction in 2005
Randall Sobie IPP/UVictoria 11
40 million collisions per second
100,000 collisions selected
200 events per second
WLCG Computing GridCERN Tier 0
10 Tier1 sites60+ Tier2 sites
140 PB data
Randall Sobie IPP/UVictoria 12
Higgs Discovery in 2012
The ATLAS and CMS experiments see evidence for a Higgs-like particle
Picture shows an event where the Higgs candidate decays to 4 electron-like particles
Randall Sobie IPP/UVictoria 13
Number of candidates (vertical axis)
Mass of the candidates(horizontal axis)
We observe an excess of candidates with a mass of
125 proton-masses
Search for Higgs decays to 4 “leptons” (electrons or muons)
Also observed in the CMS experiment
Randall Sobie IPP/UVictoria 14
July 4, 2012
15
Long-term preservation of software and data of HEP
experiments
Clouds in High Energy Physics
Randall Sobie IPP/UVictoria
Utilize special computing resources attached to the
detectors
Simplify the management of heterogeneous in-house
resources
Use commercial clouds for exceptional computing
demands
Distributed cloud computing using HEP and non-HEP
clouds
Randall Sobie IPP/UVictoria 16
Using Clouds for Data Preservation
SLAC Linear Accelerator BaBar Detector
Electron-positron collision
BaBar experiment stopped recording electron-positron collisions in 2008
Long Term Data Access system
Static SL5 VMs Full access to BaBar SW and data
Transparent to the userIn operation since 2010
http://today.slac.stanford.edu/feature/2010/babar-prototype-data-servers.asp
Randall Sobie IPP/UVictoria 17
ATLAS/CMS High Level Trigger Clouds
40 million collisions per second
100,000 collisions selected
200 events per second
HLT Farms
These systems are used in real-time when there is colliding beams
The aim is to use the resources during the idle periods for other purposes
Enabled as private OpenStack clouds
See talk by Toni Perez Wednesday 1100
50,000 cores ATLAS, CMS, LHCb
18
Private and commercial clouds
Randall Sobie IPP/UVictoria
Ibex @ CERN (J. van Eldik, T. Bell, B.Moreira)
OpenStack cloud with 5000 cores Provide batch services and cloud services
HEP using Amazon, Google, Rackspace and others
Star Experiment at RHIC (Brookhaven NL)Belle Experiment at KEK (Japan)ATLAS Experiment
Commercial clouds used for exceptional low I/O demands
Challenges: identity management, API compatibility, VM configuration and network connectivity
Costs are higher than our private resources
Randall Sobie IPP/UVictoria 19
Distributed cloud computingGrid of Clouds
Seamlessly use multiple, heterogeneous IaaS clouds
for batch workloadsSky ComputingK. Keahey etal
Use dedicated HEP and non-HEP opportunistic resources
Independent of the IaaS cloud type
Removes any application requirements from remote site
Support multiple projects
Randall Sobie IPP/UVictoria 20
Distributed cloud overview
CloudScheduler
IaaSClouds
HTCondor JobQueue
UserJob
UserVM
Discovers User-Job Boot User-VMon a cloud
Dispatches UserJobto VM
VM registers with HTCondor
Remote VM image repository (Nimbus clouds)Upload VMs to OpenStack cloudsRemote data repositories for all clouds
Randall Sobie IPP/UVictoria 21
Distributed cloud status
NimbusVictoria(3)
OttawaFutureGrid Chicago
FutureGrid SanDiegoFutureGrid Florida
OpenStackMelbourne-NECTAR
CERN-IbexCANARIE-WestCANARIE-EastOxford-GridPP
Amazon EC2Google GCE
Operational since Nov 2011
Approximately 250K jobs(Astronomy 500K jobs)
Using 10 clouds for ATLAS jobs500-1000 simultaneous jobs12 hour jobs
ATLAS jobs submitted from CERN
ç
MelbourneCERN--IbexCANARIE
WG-VictoriaNRC-Ottawa
VictoriaFG-San DiegoFG-Chicago
Victoria
April 8-15 2013100-150 8-core VMs
Randall Sobie IPP/UVictoria 22
Technology innovation for new science
The OpenStack developer community can help us
Common authentication Centralized VM image storage
Consistent meta-dataUnique cloud names
Simplify the integration of OpenStack clouds for us
We see ourselves as integrators rather than developers of cloud technology
Randall Sobie IPP/UVictoria 23
Summary
High Energy Physics
Understand our UniverseStudying the Higgs
Search for Dark MatterDifference between antimatter/matter
Impact on societyCERN is the birthplace of the WWWMedical physicsTechnology innovationHQP
ComputingLarge distributed systemsGlobal research network
Novel use of computing technologies
Randall Sobie IPP/UVictoria 24
Acknowledgements
Contact information: [email protected] sites:http://rjs.phys.uvic.ca/
http://heprc.phys.uvic.ca/home/