Se connecter

Se connecter avec OpenID

CHEP Summary - Extracts from T.Wenaus talks.

CHEP 2003 General
Torre Wenaus, BNL/CERN
CHEP 2003, UC San Diego, La Jolla
March 28, 2003
Themes and observations
Lesson from the past: Make it simple (R. Brun)
No more complex than necessary
Users want consolidation, ease of use, and stability
Must consider also needs of the future; longer view of maintainability and
In the interests of long term stability
OO and C++ is the accepted paradigm
No major OO/C++ migration or usage angst at this conference, it is done and
Offline and online: “Triumph of C++ for HEP DAQ confirmed” – DAQ
Now we are hearing reports on Nth generation C++ software
L. Sexton Kennedy, CDF: Every component has been rewritten at least
once. Implementations have now stabilized such that every new arrival
doesn’t start by discarding and rewriting software
“Many more talks about redesign than about design” – Data management
And on the maturation and emergence of tools as broad standards, after years of
development and refinement
e.g. Geant4, ROOT I/O
CHEP 2003 Summary, March 28 2003
Slide 4
Torre Wenaus, BNL/CERN
Themes and observations
The tyranny of Moore’s Law
 Wolbers: it is not a substitute for more efficient & faster code,
smaller data size
 it works against thinking before doing
 Optimize wherever possible
Addressing the digital divide in networking (H.Newman)
 HEP is obligated as a community to work on this
 A world problem in which our field can have visible impact
Farm challenges
 Don’t underestimate farm installation and operations (R.Divia)
 Big issues are power, cooling, space! (S.Wolbers)
 Watts/$ steadily rising (R.Mount)
 Tape-disk random access performance gap in analysis is receding
as an issue, but disk-memory gap is hardly being addressed
CHEP 2003 Summary, March 28 2003
Slide 5
Torre Wenaus, BNL/CERN
Rising trends
 For analysis, I/O, and much else
 Now fully supported at CERN: EP/SFT section
 Close interaction with experiments on new developments
 Run II, RHIC, ALICE, LCG, BaBar, …
 Foreign classes, PROOF, geometry, grid integration, …
 Mentioned in 47+ talks at this conference
Open source databases (MySQL, Postgres, …)
 Metadata, distributed computing, conditions, …
 Empowering software: easy and potent
 MySQL mentioned in 37 talks! Postgres in 8, Oracle in 27
Online – offline continuum
 Similar Linux farm environments, attainable time budgets
 Same framework, maybe same algorithms, in HLT as in offline
(V.Boisvert, ATLAS)
 Stringent performance/robustness requirements on software
CHEP 2003 Summary, March 28 2003
Slide 7
Torre Wenaus, BNL/CERN
Rising trends
Common projects
 Joint projects one of the CDF/D0 successes (Wolbers)
 But hard to align running experiments with LHC
 LHC Computing Grid project
 Grid projects in general
 Laudable but difficult; increasingly forced by the circumstances
 Resource constraints and increasing scale and complexity
makes go-it-alone N times too costly
 cf. comments in online/DAQ context by G. Dubois-Feldmann
today: somewhat less success in online where it is even harder
than offline, but possible LHC inroads
Related is software reuse…
 Respect what we know about long software development
CHEP 2003 Summary, March 28 2003
Slide 8
Torre Wenaus, BNL/CERN
Rene’s time to develop plot
CHEP 2003 Summary, March 28 2003
Slide 9
Torre Wenaus, BNL/CERN
LCG must effectively re-use and leverage
existing software, or fail
This is the approach taken: cf. POOL, SEAL talks. Time will tell! cf. next CHEP
CHEP 2003 Summary, March 28 2003
Slide 10
Torre Wenaus, BNL/CERN
Rising trends – The Grid
The central importance of distributed computing to future
(increasingly, present) HENP is long known
‘The Grid’ as the means to that is now established
 Major, broad successes in funding and in attracting
collaboration with CS
 F.Berman, Grid 2003: “HEP has set a model for
integration, focus, coordination”
 Progress in applying Grid software and infrastructure
to real problems
 Batch production
 Clearly the chosen path; success to be proven, but has
promise and broad commitment
CHEP 2003 Summary, March 28 2003
Slide 15
Torre Wenaus, BNL/CERN
The Grid
F.Berman, Grids on the horizon:
 Must be useful, usable, stable; supported
 More cooperative than competitive
 [Not always the case today!]
 Applications are key to success
 Not a “Field of Dreams” “build it and they will
come” R&D field any more
 Grid killer app: a focus on data. Good match to us
 Still a long way to go
CHEP 2003 Summary, March 28 2003
Slide 16
Torre Wenaus, BNL/CERN
The Grid
Miron Livny:
 Benefit to science: democratization of computing
 Still very manpower intensive: when the support team
goes on holiday, so does the Grid (CMS testbed in Dec)
 Best practice middleware requires
 True collaboration, “open minds” (cf. Berman)
 Testing, deployment/adoption, evaluation metrics,
robustness, professional support, longevity,
responsiveness to show stoppers, …
 Much to do and improve but important progress
 E.g. VDT as standard middleware suite
CHEP 2003 Summary, March 28 2003
Slide 17
Torre Wenaus, BNL/CERN
Receding trends
Objectivity and ODBMS in general
 “Jury still out” at CHEP 2000 (P.Sphicas), but now clear
 Objectivity dropped or being phased out by LHC experiments,
COMPASS, BaBar event store
 In PHENIX “becoming a liability” (compiler issues);
augmented with RDBMSs
 Not due to technical failure but a mix of technical problems,
commercial concerns, manpower costs, availability of an
 Its replacements are not other ODBMSes but files (often ROOT)
+ RDBMS (mySQL, Oracle, Postgres…) for metadata
Magnetic tape (apart from archival)
 PASTA: “unlimited” multi-PB disk caches technically possible
but true cost is unclear (reliability, manageability)
 File system access under urgent investigation
 “tapes as random access device no longer a viable option” –
large disk caches needed for LHC analysis
CHEP 2003 Summary, March 28 2003
Slide 22
Torre Wenaus, BNL/CERN
Receding trends
Commercial software? No…
 Some in decline (Objy, LHC++), but new prospects
opening (IBM, Sun, MS, …) in Grid
 Open source now has an important commercial
element we derive great benefit from (even
 Red Hat, MySQL, Qt, …
CHEP 2003 Summary, March 28 2003
Slide 23
Torre Wenaus, BNL/CERN
Collaborative tools
 Was represented this week, but only lightly
 Vital for distributed collaboration on software development and
physics analysis
 H. Newman: need culture of collaboration
 Distributed and remote collaboration should be the norm
 Not solely, or even predominantly?, a matter of tool development
in the community
 How is the exponential commercial side evolving and how
can we leverage it
 What is the evolutionary path, strategy, role for communitydeveloped tools such as VRVS
 Why is the user experience often poor
Poor physical facilities/configurations, instabilities,
heterogeneous tools/protocols, support issues, …
Current experience sometimes competes unsuccessfully with the
telephone, despite all the shortcomings
CHEP 2003 Summary, March 28 2003
Slide 24
Torre Wenaus, BNL/CERN
Data analysis as “the last wheel of the car” (R. Brun)
 Clear message from current generation (e.g. Run 2, BaBar): don’t
leave data analysis systems and infrastructure too late, it will lead
to problems
 Vastly more true when we are talking about doing globally
distributed analysis, for the first time
 with unprecedented volume and complexity, e.g. Terabyte
scale at the LHC
Making dist analysis both very difficult and mandatory
We cannot bootstrap ourselves into a global analysis system, it
will take long incremental work, so we better be working in a
coordinated & effective way now
R. Brun: Will not converge on one system; will be multiple
competing systems, and that will not be bad [hopefully a small
CHEP 2003 Summary, March 28 2003
Slide 33
Torre Wenaus, BNL/CERN
Are we doing enough to ensure senior people can
contribute directly to physics analysis?
 How do we interpret the fact (R. Brun) that PAW usage
is still rising?
 Has everyone bought the C++/OO paradigm shift?
 Are we developing and/or providing the right tools?
Is there enough engagement of senior physicists in the
(limited) exploratory work being done on future physics
analysis environments?
 Almost certainly no, and may be difficult to attract
their attention unless/until attractive prototypes can be
turned loose on them
CHEP 2003 Summary, March 28 2003
Slide 34
Torre Wenaus, BNL/CERN
Conclusions (2)
Grids and networking are making great strides
 HENP is a successful and valued partner with CS
 We provide a community focused on challenging
large-scale deployments in real research settings
 But Murphy’s Law is a potent adversary today; far
from robust transparency, and much much more to do
Global collaborative computing must become a successful
norm for us
 Down to the global researcher at the home institute
 Rich leadership potential for our field
Important new common endeavours like the Grid and
LCG have much invested in their success… will be
interesting to measure the degree of success at next CHEP
CHEP 2003 Summary, March 28 2003
Slide 38
Torre Wenaus, BNL/CERN
Без категории
Taille du fichier
148 Кб