The Software AG product ETS Optimize for Infrastructure enables you to monitor all Software AG component resources in real time. A global buffer pool, in the following called Optimize Monitor Buffer Pool, is available in Natural to provide system and operational data for monitoring all Natural components running in one LPAR (z/OS), on one machine (z/VSE) or on one host (BS2000/OSD).

This part describes the use of the Optimize Monitor Buffer Pool.

The following topics are covered:

• Purpose

• Prerequisites

• Starting the Optimize Monitor Buffer Pool

• Operator Commands

• Messages

---

Purpose

All Optimize for Infrastructure Key Performance Indicator (KPI) data is collected in an Optimize Monitor Buffer Pool per LPAR (z/OS), machine (z/VSE) or host (BS2000/OSD). This buffer pool actually is a data space.

It consists of the following parts:

• Pool Header

  The pool header holds control information and statistical data.

• KPI Pool

  The KPI pool holds all instances with their KPI data slots.

• Session Data Pool

  This is the pool where all sessions put their session data per dialog step.

  The session data pool is written to in a wrap-around manner. Session data is aggregated by a back-ground task (aggregation daemon) into the associated KPI slots.

Prerequisites

The following prerequisites must be met if you want to use the Optimize Monitor Buffer Pool:

1. The Optimize Monitor Buffer Pool must have been created and started; see the corresponding step in Installing Natural on z/OS in the Installation documentation.

2. The profile parameter O4I must be set to ON.

Starting the Optimize Monitor Buffer Pool

You start the Optimize Monitor Buffer Pool either as a batch job or as a started task by executing module NATO4Ivr (where vr represents the relevant product version).

To start the Optimize Monitor Buffer Pool

• Specify the following parameters as PARM in the JCL EXEC statement (z/OS and z/VSE) or in the SYSDTA logical system file (BS2000/OSD):

  bp-size,subpool-size,daemon-idle-time,filled-threshold

  All parameters are positional and must be separated by a comma. They are explained in the table below:

  bp-size	Total Optimize Monitor Buffer Pool size in KB. Can also be specified in units of MB or GB by adding a trailing M or G at the end of the size value.
  subpool-size	Size of KPI sub-pool.
  daemon-idle-time	Aggregation daemon idle time in seconds. Valid Values:
  	1 - 60 This is the time interval the aggregation daemon will regularly wake up and aggregate the session data into KPI data. 30 This is the default value.
  filled-threshold	Session data pool "filled" threshold, when the aggregation daemon is triggered for immediate wake-up. Valid values:
  	20 - 80 "Filled" threshold in percent. 50 This is the default value.

Operator Commands

Notation:

The operator commands can be specified in any length from minimum (capital letter) up to full command name length.

End

Terminates the Optimize Monitor Buffer Pool.

In z/OS, the Optimize Monitor Buffer Pool can also be terminated via the operator command STOP.

Halt

Stops the aggregation daemon.

Start

Restarts the aggregation daemon.

When restarted, the aggregation daemon starts from session data pool start.

Idle=nn

Sets the aggregation daemon idle time. nn = value in seconds.

Trigger=nn

Sets the aggregation daemon trigger threshold. nn = value in percent

Messages

Refer to Optimize Monitor Buffer Pool Messages in the Messages and Codes documentation.