STAR   Computing Tutorials main page
DataBase & Server Layout
Offline computing tutorial Maintained by Jeff Porter
Last modified Fri Jul 03 00:19:00 1998

Introduction

There are a set boundary conditions which dictate the layout for databases and the servers though which read/write access is granted. These are,
  • Online & Offline require functional autonomy
  • Write access is controlled at the "subsystem" (Online) or "domain" (Offline) level.
  • MySQL provides simple (write) access control on a per database basis.
  • While allowing copies (mirrors), data integrity requires 1 (& only 1) primary sink/source per DB
  • Most database types have a well defined write clients in terms of Online & Offline : See list on DB-Types
A simple schematic is shown here.


Online-Offline divisions

Below is a simple table to show the where the primary DB data source for the following database types. The term Primary Source refers to where write access is granted.

DataBase TypeDomain List Primary Source
Conditions Daq,Trg,Tpc,Emc,... Online
Configurations Daq,Trg,Tpc,Emc,... Online
Scalars Trg, RHIC Online
RunLog Daq,Trg Online
Calibrations Tpc,Emc,... Offline
Geometry Tpc,Emc,... Offline
Run-or-Code Params Tpc,Emc,... Offline
TagDb ? Offline
Production None Offline


Navigation Structure

Below is a table of how one navigates (finds instances) within databases. This table corresponds to the "flow chart" and index specifications noted on the Navigation page.

DataBase Type Index Types
Conditions TimeStamp
Configurations Key and Version for initializing detectors, TimeStamp for later reference
Scalars TimeStamp
RunLog Run & Event number
Calibrations version and TimeStamp
Geometry Version (or TimeStamp-map)
Run-or-Code Params Version (or TimeStamp-map)
TagDb Run & Event number
Production Run, Event number, production Index