Summary of Computing Section of Technical Proposal 9292020

Summary of Computing Section of Technical Proposal • • • 9/29/2020 Data Flow Model Computing Requirements Computing Infrastructure Software Strategy Project Organisation and Management Manpower estimates and costs Computing section of LHC-B TP 1

Status of Documents r Draft of Computing Section is available - 5 pages r Based on four Computing Notes containing more details ã ã LHC-B Computing Tasks Requirements LHC-B Computing Model LHC- B Software Strategy LHC-B Project Plan for Computing r Drafts of notes available on Web - still being revised 9/29/2020 Computing section of LHC-B TP 2

9/29/2020 Computing section of LHC-B TP 3

Data Flow Model r Algorithms used for Level 2/3 triggers similar to those employed in full reconstruction. Issues are : ã speed, reliability ã calibration and alignment in real-time ã output of L 2 and L 3 used by full reconstruction r Size of data store and access speeds 2 -3 orders of magnitude higher than current experiments and similar to other LHC experiments ã ã 9/29/2020 Raw data written to storage at 20 MB/s. Similar amount of reconstruction information (14 MB/s) In total capability of storing > 0. 4 PB of data/year at 40 MB/s Transparent access to data store by nearly all tasks Computing section of LHC-B TP 4

Computing Requirements r Estimates of CPU requirements, input/output data volumes based on… ã simulation - program exists so good estimates m Assumptions on evolution of algorithms : optimisation (e. g. shower parameterisation) – increasing complexity (more detail) : new frameworks like GEANT 4 (30% improvement) ã reconstruction - partial information on pattern recognition m Extrapolations from existing experiments (need input from HERA-B) ã analysis algorithms - less well known but needs are smaller r Some numbers are ‘targets’ as opposed to ‘benchmarks’ ã for example, goals for L 2/L 3 are 10/200 msec (on 1000 Mips CPU) 9/29/2020 Computing section of LHC-B TP 5

Dedicated Installed Processing Power 9/29/2020 Computing section of LHC-B TP 6

Data Storage Requirements 9/29/2020 Computing section of LHC-B TP 7

Computing Infrastructure r Issues are ã ã ã 9/29/2020 Strategy for evolution of computing model Timescales for investment in computing resources Scalability of cpu farms needed for cpu intensive processing Handling of Petabytes of data stored in a central database Equal access to data for all collaboration institutes Computing section of LHC-B TP 8

Evolution of Computing Infrastructure r Steady investment in desktop systems r Preparation Phase (1998 - 2000) ã ã 1998 - need is 1000 Mips and 2 TB of data Use public facilities both inside and outside CERN Increase of 50%/year in requirements for simulation and analysis Impact of test-beam? r Implementation Phase (2001 - 2003) ã significant increase in our needs (TDRs, full MC, testbeam) ã invest in private (SHIFT-like) facilities (end 2000/beginning of 2001) r Commissioning Phase (2004 - 2005) ã assembly of full-scale facilities ã financing scheme 9/29/2020 Computing section of LHC-B TP 9

Data Storage Model r r All event data stored in a single Object Database Storage/retrieval managed by a hierarchical mass storage system Assume 10% of data stored on disk Study options for access of data from any institute ã ã 9/29/2020 CERNtric model - all data stored at and accessed from CERN Regional centres - data distributed between CERN and home labs Cache (part of) data at each institute Depends on technology (network), tariffs, logistics, politics Computing section of LHC-B TP 10

Software Strategy r Objectives ã quality in software ( trigger, prompt reconstruction…. ) ã performance - trigger latencies, CPU for bulk processing ã improve on : m knowledge of PEOPLE involved m the organisation of the development PROCESS m the TECHNOLOGY used r Approach ã ã ã ã 9/29/2020 use appropriate engineering practices stress importance of architecture - adherence to standards build high quality components (manpower intensive) re-use components wherever possible (manpower efficient) use commercial products when appropriate participate in common (LHC-wide) projects plan well - encourage all members of collaboration to participate Computing section of LHC-B TP 11

Software Strategy Technology r Specialised tools that help building software for all life-cycle activities ã project management (MSProject, communication (web), workflow) ã verification (inspection, testing) - Purify, Logiscope ã configuration management (code and documents of all sorts) r Technology for life-cycle phases ã TP states “our intention is to adopt Object Technologies” ã OOA (analysis), OOD (design), ODBMS (database), C++/Java (language) integration standards (OMG/CORBA, Active. X/DCOM, RMI/Javabeans) ã large investment by software industry - commercial tools and products widely available (GUIs, distributed systems) ã widespread adoption within HEP m GEANT 4 - new simulation framework re-engineered using OO m Event Store/Objectivity m Replacement of CERNLIB - Open. GL, Iris. Explorer (analysis framework) m Adoption by other experiments (Ba. Bar, STAR, ATLAS/CMS, ALICE. . ) 9/29/2020 Computing section of LHC-B TP 12

Benefits of OO r OO evolved out of addressing issues of “programming-in-the-large” r Objects are basis for reusable modules r Communication by message passing helps to define interfaces between modules and external systems r Design essential features of an object that distinguish it from all other objects - defines crisp boundaries (Abstraction) r All internal implementation details are hidden - manage complexity (Encapsulation) r Reuse of well designed/tested modules (objects) gives better quality and leads to high productivity r Partitioning of work into domains is much easier 9/29/2020 Computing section of LHC-B TP 13

Drawbacks of OO r Field is still developing rapidly and some technologies/products may be superceded r Culture change is necessary and , in general, people hate this r Significant costs associated with training and re-education r OO may not be the last word in software engineering 9/29/2020 Computing section of LHC-B TP 14

Migration Policy r Steps are as follows : ã Build up a suitable programming environment (e. g. C++, UML, Rose) ã Develop frameworks for simulation, reconstruction and analysis m impetus will come mid ‘ 98 with release of GEANT 4 and LHC++ toolkits ã Embark on intensive training programme r Minimise legacy software - hence set an aggressive schedule r Manpower is an important issue ã consolidation of SICB development ã need extra (skilled) effort 9/29/2020 Computing section of LHC-B TP 15

Steering Group • Composition - coordinator plus one rep from each project • Tasks - Coordination, Planning, Resources Computing Facilities • Farms • Desktop • Storage • Network • Operating System Recon- Analysis Simulation DAQ struction • Level 2 FW • Framewk • Level 3 FW • Tools • Recon FW • Calibration • Production • GEANT 4 Framewk • Tools • Production Controls OPS Software Eng. Group • DCS • LHC • Safety • Run Control • Operations • Consoles • Shift Crew Enviroment • Methods • Tools • Code Manag. • Quality • Document. • Training • Licenses • Collab. Tools • Event Builder • Readout Network • Interfaces • Links • Crates • DAQware Re-usable Components • Data Management : Event Store, Geometry, Database Utilities, ODBMS • Architecture : Frameworks, Component model, Distributed system • Toolkits : GUI, Histograms, Communications • Utilities : data quality monitoring, event display, bookkeeping 9/29/2020 Computing section of LHC-B TP 16

Links to Sub-detector Groups Application Project (e. g. Reconstruction) • Project Leader • Vertex • RICH • Inner Tracker • Outer Tracker • ECAL • HCAL • MUON • Trigger L 0 • Trigger L 1 • Trigger L 2/L 3 9/29/2020 RICH Computing Team Computing section of LHC-B TP 17

Life-Cycle Phases r Preparation Phase ( now until end of 2000) Learning ã collect requirements and develop functional specifications of subsystems ã evaluate hardware technologies ã build prototypes r Implementation Phase (start ‘ 01 until end ‘ 03) Building ã make technology choices ã engineer sub-systems r Commissioning Phase (start ‘ 04 until end ‘ 04) Testing ã install ã unit test, integration tests ã tests under realistic loads (bulk data, realistic real-time tests) r Operation Phase (start ‘ 05 until physics goals archived) Running ã support ã adapt and improve 9/29/2020 Computing section of LHC-B TP 18

Manpower Estimates Group 98 99 00 01 02 03 04 05 Steering Group DAQ Controls Operations Simulation Reconstruction Analysis Re-usable components Software Engineering Computing Facilities 1 4 1 0 3 2 2 2 1 6 1 0 3 2 2 2 10 2 0 3 3 3 7 5 5 2 10 3 1 3 3 4 7 5 5 2 10 3 2 2 3 4 4 4 8 2 4 2 3 4 4 4 8 TOTALS 19 21 22 41 43 42 41 9/29/2020 40 Computing section of LHC-B TP Comments + 1 -2/subdetector Common Project + 1 -2/sub-detector interactive applications Common Project 19

Cost Estimate Initial Investment Cost Item Units Unit Cost 1 Total Cost CPU (Mips) 1 x 106 3 SFr 3. 0 MSFr Disk 2 (TB) 42 12 k. SFr 0. 5 MSFr Tape (TB) 420 1 k. SFr 0. 5 MSFr Total 4. 0 MSFr Notes : 1. Taken from industry supplied extrapolations to the year 2005 2. Assume 10% of total data taken will reside on disk Item Desktop CPU Software (LHC++, OS) CPU Disk Tape Annual Investment Costs Unit Cost 100 SFr/month Total 9/29/2020 Computing section of LHC-B TP Total Cost 100 k. SFr 500 k. SFR 200 k. SFr 500 k. SFr 1400 k. SFr 20

9/29/2020 Computing section of LHC-B TP 21