ACAT 2010 Axel Naumann

What? International Workshop …on Advanced Computing …and Analysis Techniques …in Physics Research "Beyond the Cutting edge in Computing" February Jaipur, India, at University of Rajasthan Axel Naumann - SFT2

Content 1.Computing Technology for Physics Research 2.Data Analysis - Algorithms and Tools 3.Methodology of Computations in Theoretical Physics And "hot topics", this time: multicore data management 50% plenary, 50% parallel, 2 hot topic panels Axel Naumann - SFT

Location Incredible India Axel Naumann - SFT

Location Incredible People Axel Naumann - SFT

Location Axel Naumann - SFT6 New Delhi Jaipur

Participants About 80 participants Far away: really interested participants Few new people: family But big family, e.g. – CERN, e.g. all main LHC experiments (LHCb?) – Fermilab, BNL, SLAC,… – KEK (Belle II); of course India SFT: Rene, Fons, Gerri, Matevz, Axel Axel Naumann - SFT7

SFT (related) Plenaries Rene: "History of the ROOT System: Conception, Evolution and Experience" Matevz's lecture "Simulation and Visualisation Techniques" Axel Naumann - SFT8

Rene's Plenary Status of ROOT in 2010; future Historical background of alternatives, technological and sociological decisions "Lifetime achievement award" Axel Naumann - SFT9

COMPUTING TECHNOLOGY FOR PHYSICS RESEARCH Languages, Software quality, IDE and User Interfaces Distributed and Parallel Computing New architectures, many and multi-cores Virtualization Online Monitoring and Control Axel Naumann - SFT

SFT Contributions Fons: PROOF - Best Practices Always fighting bottlenecks – file layout: much improved – merge: multiple mergers – RAM – CPUs: PROOF lite – disks, network: comparing storage solutions Feeding 24 cores not trivial Axel: checking coding rules with LLVM's clang Axel Naumann - SFT

Clouds From theory to practice to integration – Use cases still investigated – How to justify local batch vs. paying cloud Why Clouds == outsourcing? – Burst resource on peak demands? – Opportunistic resources truly at reach? Understand virtualization. Still open: – is it needed? do we want it? – is it transparent enough for end-users? – integrity, reproducibility? – image management, networking? Axel Naumann - SFT

GridLite Usability Management Fights are over, reality arrived 1/3 of jobs lost: inherent property of the grid? Mattia Cinquilli Axel Naumann - SFT

Multi- & Other-Cores x86_64 understood, optimizing An occasion to optimize / re-think our code Finally again competition across chips designs / architectures! What will be the surviving architecture? – Lots of dynamic … Axel Naumann - SFT

Tools & Languages Many tools, only some mastered. Do we need new non-expert tools? – New compilers for parallelization – New tools for detecting language constructs / best practice / code analysis – Library / extension helping novice to develop parallel applications (OpenCL) Languages vs. optimizations / architectures: do we sell our souls? Standards! C++ 0x not a topic yet, unlike 64bit CPUs when they came out, Larrabee,… Axel Naumann - SFT

Data Equilibrium money - CPU - I/O: alchemy CPU focus swinging to I/O: did we overestimate relevance of CPU (interpreter in event loop)? Cache hierarchy: tape, disks, SSDs, RAM – All of the above? – Throw a WAN in? – Need algorithms for data lifetime across tiers Latency vs. cooperation / concurrency, theory! 24 core (“wow!”) but no data coming through! Axel Naumann - SFT

DATA ANALYSIS – ALGORITHMS, TOOLS Machine Learning Advanced Data Analysis Environments Simulation, Reconstruction and Visualisation Techniques Advanced Computing Axel Naumann - SFT

Data Analysis – Algorithms, Tools Patterns Axel Naumann - SFT18

Data Analysis – Algorithms, Tools Parallelization of reco algos Still exploring new algos for data analysis Evolutionary algos * quantim computing Code generators for data analysis Lots of ROOT; special presentations on TMVA, RooStats Axel Naumann - SFT19

COMPUTATIONS IN THEORETICAL PHYSICS: TECHNIQUES AND METHODS Automatic Systems Higher orders Computer Algebra Techniques and Applications Axel Naumann - SFT

Methodology of Computations in Theoretical Physics Recursion Axel Naumann - SFT21

Methodology of Computations in Theoretical Physics Very good plenary by Daniel Maître: "Automation of multi-leg one-loop virtual amplitudes" Fortran rules Mixing analytical and numerical approaches Numerical instabilities (>= quad precision) Axel Naumann - SFT22

Conclusion Real workshop atmosphere Very productive discussions resulting in real projects! Small is beautiful (but not sufficient) Everybody took something back Axel Naumann - SFT

References (Almost) all presentations in Indico: Axel Naumann - SFT24