SLAC PEP-II
BABAR
SLAC<->RAL
Babar logo
HEPIC E,S & H Databases PDG HEP preprints
Organization Detector Computing Physics Documentation
Personnel Glossary Sitemap Search Hypernews
Unwrap page!
Simulation Home
Sim Codes
Event Generators
Bogus/BgsApp
SimApp
Bear
Moose
Fast Simulation
Geant4 Home
Subsystems
PEP
SVT
DCH
DRC
EMC
IFR
Mixing/Trigger
Backgrounds
Mixing
Trigger Simulation
MC Truth/QA
MC Truth
Micro/Mini
QA Histograms
Sim Error Reports
REMEDY
MC Production
Production Home
Test Production
Tools
Database
CERNLIB
CLHEP
Event display
RandControl
Scripts
Check this page for HTML 4.01 Transitional compliance with the
W3C Validator
(More checks...)

Schedule Coupling

These notes are an attempt to explain the coupling between the
Release Schedule and the Production Schedule.

o Release deadline has been moved to Monday. This is the day on which
building of a release commences.

o There is a release deadline every two weeks. It is foreseen that there
should be a serious attempt to encourage package coordinators to
distinguish between two types of release which alternate. These are:
  - MDC "bug-fix only" release
  - new functionality release
Of course, this is hard to impose, but it should at least be encouraged.
Clearly each type of release happens only once per 4 weeks.

o In the past, building a release has taken typically 4-7 days. Production
should therefore allow 7 days before they can reasonably expect a new
release to be announced.

o Following announcement of a release, it is anticipated that the
Production Team spend several days testing it. Bugs will be reported to
the Simulation and/or Reconstruction managers, who will organise the
appropriate individuals to fix the bugs in their packages. On tagging
of bug-fixed packages, the release team (Colorado) will rebuild the 
executables (with lower-case letter subscripts appended to release
number of executable). Schedule-planning should allow at least 7 days for
this process to take place. In the short term, it might be expected that
the simulation executable would be ready before the reconstruction one, 
as it is much more stable, a priori. This matches well with the logical
processing order, allowing simu to begin before reco is fully debugged, 
if necessary.

We therefore propose the following coupled schedule. Dates are not fixed
in stone, but represent planning guidelines.

         Day
Monday 1   1  Production Release deadline (these are already foreseen in
                                          the published release schedule).
Monday 2   8  Executables should be ~ready. Release announced. 
              Production testing begins.

Monday 3  15  Production expect to begin large-scale simulation. 
              Release deadline for new functionality release.

Monday 4  22  Production expect to begin large-scale reconstruction.

Monday 5  29  Start of next cycle = Day 1