Classic event subsystem performance factors

The overall performance of the AVEVA Historian Classic Event subsystem is subject to factors related to data storage and query processing time. Too often, systems are commissioned with specifications that estimate average or "typical" expected loading. Instead, you should size the system so that it can accommodate the peak load that you expect during the projected system life cycle. Some performance factors you should consider are:

• Sufficient hardware.
  Your selection of hardware is important to guarantee peak performance for the range of behaviors required for a given operating environment. For example, you should make sure that you have enough disk space to store the records of detected events and the results of any actions (summaries, value snapshots, and so on).

• Processor availability.
  The Classic Event subsystem is subject to processor availability as much as any other software sharing a common platform. At any given moment, multiple processes contend for processor time.

• Nature of the database queries executed by the Classic Event subsystem.
  For example, because Classic Event subsystem actions typically operate on normal SQL Server tables, they are subject to performance limitations of the Microsoft SQL Server. Also, query activity tends to be very CPU-intensive and is extremely sensitive to other concurrent activities being performed on the same server.

• Time intervals for SQL-based detectors.
  For more information, see Time intervals for SQL-based detectors.

Performance can vary greatly for the same event task, depending upon the computer configuration, user interaction, and other unpredictable activity common in a plant situation with shared database and server resources. It is often very difficult to determine precisely what combinations of hardware and software parameters will work optimally for your required operating environment. Therefore, you should test your Classic Event subsystem configuration before running it in a production environment to make sure that the system will not become overloaded during peak use.