Software Status

Murrough Landon - 16 April 2002
http://www.hep.ph.qmul.ac.uk/landon/talks

Overview

• Effort
• Packages
• Test vectors
• Next steps
• Schedule

Effort

People

• Heidelberg: Oliver is leaving in August.
• Mainz: new PhD student (Michael Bussmann) starting in June (for one year, then he goes to D0).

Distractions

• Cabling document!
• Task forces and working groups (ROD crate DAQ, Online software requirements reviews, condition database)
• Calibration discussions

Packages (1)

HDMC

• VME driver for Heidelberg ``home brew'' CPU.
• Our old VME drivers dont work under RedHat 7.2 (kernel 2.4). Now using new VME driver from CERN (for Concurrent CPUs).

Module Services

• Almost complete?
• Apart from documentation...
• Move to CMT for common build process?

Packages (2)

Databases

• Not much development recently.
• API of integrated database objects must be finished to allow progress with CPM and CMM.
• Further development for overall management of test vector runs is still requried.

Run Control

• Still required: DSS ``kicker'', calibration/test sequencer and starting those processes automatically.

Packages (3)

Simulation

• Work started on CPROD simulation.
• Integration of Bill Stokes test vector generator.
• CMM simulation still required.
• Still required: CMM simulation, more integration with database, move to CMT (desirable).

Test Vectors

• Still need code to generate test vectors for CMM.

Test Vector Organisation

• Largely a database issue: named collections of run types with configuration and calibration data, sets of test vectors and simulated outputs, etc.
• Another apsect is L1A handling. Solution using DSS agreed and DSS firmware changes specified.
• Recent software discussions reviewed the above. But more details need to be defined, agreed and implemented.

Next Steps

• Complete work on database, fix APIs
• Complete CMM, CPM module services using database objects
• Develop CMM and CPROD simulation
• Implement ideas on organising test vectors
• Integration(s) of existing modular components

Timescale (1)

Optimistic estimates? [At Heidelberg!]

• Database work: 4 weeks?
• Module services work: 2 weeks? + documentation
• Test vector scheme: 4 weeks?
• Module services for CPM, CMM: 2-4 weeks each?
• Simulation of CMM: 2 weeks?
• More test vector generators: few weeks?
• Run control related: 2 weeks?
• Various integration tests: 2-4 weeks
• Not likely to be ready before June??

Timescale (2)

Overall Schedule

• UK software work until end of June
• Followed by software and hardware testing
• JEP (DAQ) software integration after that
• Followed by preprocessor...