Performance tuning
Tivoli Netcool/OMNIbus performance can be measured in terms of response time, throughput, and availability.
These performance metrics are affected by several factors,
which include:
- The number of events, including details and journals, in the ObjectServer
This number has a direct bearing on the volume of data that is transferred during resynchronization operations, and on the number of rows that need to be scanned during SQL queries from clients and triggers.
- The event throughput into the system from probes or inbound gateways
- The event throughput out of the system from outbound gateways
- The number of concurrent Web GUI and desktop display clients, and the number of other connected clients that compete for the ObjectServer's time
- The number and complexity of Web GUI and desktop filters, and the number of connections and complexity of actions in the policies that are submitted by Netcool/Impact clients, where applicable
- The efficiency of the custom ObjectServer automations, and whether they are subjecting the ObjectServer to unnecessary workload
- The granularity settings of the ObjectServer (that is, how often it sends IDUC broadcasts to clients)
- Your hardware and system configuration
To help obtain an optimal level of performance for the ObjectServer, you can monitor the number and throughput of connecting clients, evaluate the efficiency of the ObjectServer automations, and design efficient indexes to support your SQL queries. Also consider the use of a multitiered architecture to support high availability of your Tivoli Netcool/OMNIbus installation.