70 likes | 81 Views
This collaborative document outlines the planning timeline for the SuperB computing project, defining R&D efforts, setting goals for the main detector and collider TDR, and providing directions for computing R&D in the near to medium term. A comparison with the BaBar timeline is included, highlighting key activities and milestones for both projects.
E N D
SuperB computing planning Gregory Dubois-FelsmannFabrizio BianchiDavid BrownPeter ElmerMauro MorandinRoberto Stroili
Scope of present planning work • Define • A notional timeline for the development of the SuperB computing project • Goals for what should be included in the main detector and collider TDR with respect to computing, and what should be deferred to an additional computing-specific TDR that might be published a year or two later than the main TDR • Directions for computing R&D for SuperB in the near to medium term
Comparison with BaBar timeline • We have reviewed the extent to which the BaBar and SuperB timelines can be matched up. Assuming first collisions in 2015, corresponding to 1999 for BaBar: BaBar year BaBar activities SuperB year 1992 Blue-sky computing R&D 2008 C++ experiments (e.g., gismo) ASLUND and other Fortran simus for det. design 1993 Project start / collaboration formed 2009 TDR-writing 1995 RJ appointed reco manager / rough design 2011 freeze for overall computing design 1996 Beta appears; physics workshops start; MDC1 2012 Serious online software development starts 1997 Last really major changes to offline design 2013 1998 MDC2; online software supports commissioning 2014 1999 First collisions 2015 2003 CM2 2019
Proposed project phases • Four-phase timeline for SuperB based on this analysis: - Now - • Phase 0: our present work; definition of R&D plan; report to an all-hands SuperB workshop mid-2009; continueddevelopment of simulation tools - mid-2009 - • Phase 1: major R&D program for SuperB computing; invitation to new people and new ideas; clean-sheet redesign of Beta and selected portions of BaBar s/w (e.g., DIRC reco); continued support and development of simulation tools for TDR and subsequent detector development; probable physics-book-type activity • Designing for performance is a major cross-cutting goal; training available for project participants • Formal SuperB project start sometime during Phase 1 • End of phase 1: computing TDR complete or largely so
Proposed project phases, continued - mid-2011 - • Phase 2: integration of R&D program results into a complete SuperB software system; major online software development gets under way; series of data challenges • End of phase 2: fully-functional beta of SuperB offline computing, usable for physics studies; retirement of the phase-0 simulation tools - end of 2013 - • Phase 3: scaling tests and development; converging on final full-scale system; acquisition of hardware; use of online system for detector integration, cosmics, commissioning - 2015 - • First beams
Emphasis on performance • We have agreed on a shift in emphasis toward designing for performance at an earlier stage of the project than was feasible in BaBar. • We also agree on the importance of making available to physicists training in designing for and measuring performance, so that the burden does not have to be carried entirely by computing professionals.
Current work of the group • Identify areas for R&D efforts during Phase 0 and Phase 1