[ Home | Members | Detector | Electronics | Trigger | Online | Database | Measurements ]
Database
Luminosity Recording Requirements
The D0 Luminosity is based upon our best estimate of the collider luminosity. This luminosity is calculated from the Luminosity Monitor counting rate using the inelastic and diffractive cross sections, the Luminosity Monitor acceptance and efficiency, and multiple interaction corrections. Due to the non-linear nature of the multiple interaction corrections, it is necessary to track the luminosity for each of the beam bunches.
The collider luminosity is then corrected for non-physics effects to obtain the D0 Luminosity. These corrections include:
The D0 Luminosity does not include corrections for:
The above effects will depend on the physics process and will require work by the physics groups to make the appropriate corrections. Physics groups are urged to provide sufficient monitor triggers to enable the above corrections.
The online measurement of the D0 Luminosity will be updated at least once per minute. The luminosity will be calculated separately for each beam bunch for each of the deadtime conditions defined by the trigger. A delivered luminosity that reflects the collider luminosity will also be calculated. The integrated D0 Luminosity will be calculated for each trigger condition and recorded in a database at least once every five minutes. In addition, the integrated D0 Luminosities will be recorded upon the occurrence of a significant event, such as pausing/ending a run or a sudden decrease in the luminosity.
All information needed to calculate the luminosity will be stored with the luminosity scalars in the same database. This includes the trigger names, relationships between L1, L2, and L3 trigger bits, the prescale factors, the appropriate deadtime measurement for each trigger bit, and time stamps for starting and stopping data recording. In addition, it is expected that there will be alternate luminosity measures that will need to be recorded to provide cross checks on the primary luminosity measurement.
Since we need to provide an accounting of how the delivered luminosity was used, the database should facilitate correlating the various sources of deadtime and downtime with the associated loss of luminosity.
Each file produced by the offline reconstruction and streaming programs will have the integrated luminosity for each relevant trigger recorded in a database. These luminosities will be corrected for events that are discarded in the reconstruction and streaming jobs.
Programs will be provided to extract the luminosity for a given trigger from a set of files used in the analysis. These programs should be easy
to use and quickly return the desired luminosity.
If we adopt exclusive streams, there may be situations where two or more streams are sufficiently correlated that a missing file in one stream requires exclusion of the corresponding data in the other stream(s) to properly account for trigger efficiencies. A method will be provided to find the luminosity with such exclusions.
