Introduction

Run logging is the process of collecting run dependent information into a series of banks and storing them in the database for future reference. The information is collected from the data stream and comes both from the permanent banks, such as the

RHDR bank

and the

TRIG bank

and the event data, i.e. the

ZDAB bank

Information from the input banks is summarised at different levels. At one extreme, banks such as the TRIG are kept in full, whilst at the other, the PMT data from the ZDAB banks are just held as histograms. In between these two, the position of the manipulator is held in an n-tuple. HBOOK uses ZEBRA banks to store n-tuples and histograms so they can be transferred intact to the RLG data structure. However, this means that parts of this structure are are only suitable for processing by HBOOK. To provide support for the collection and analysis of runlogger data, 4 processors are provided:-

RLG: Runlog Generator

This processes the entire data stream for a single run and produces an RLG data structure that is written to the database.

RLR: Runlog Reader

This can read, from the database, a set of RLG data structures in a selected run or date range, or drawn from a run list. If available the RLAI structure is also read and attached to the RLOG structure. It can also generate the associated DAMM histograms.

RLO: Runlog Output

This is used with RLR to output the HBOOK data (histograms and n-tuples) into a user specified HBOOK file.

LSP: Livetime Summary Processor

This is used with RLR to provide a livetime summary for the input runs.

In addition to these 4 processors there is also one RAA subprocessor:-

RAA Runlog Analysis

This produces the RLAI bank and its sub-banks. These are designed to hold data that may change from time to time, but avoids the need to recreate the entire runlog structure.