Operations Management on premises data flow
Use this information to understand how event data is retrieved from a monitored application environment and transferred between the products and components of the base Netcool® Operations Insight® to provide Event Analytics and Event Search capabilities.
The following figure shows a simplified data flow between the products of the base Netcool Operations Insight solution.
The stages of this data flow are as follows, indicated by the callout graphics (for example, 1 ).
- Capture of alert data
- Probes monitor the devices and applications in the environment.
- 1 : Alerts are received from applications and devices
- Alert data is captured by the probes and forwarded to the Netcool/OMNIbus ObjectServer. Event data is then manipulated in various data flows.
- Web GUI data flow
- Event data is enriched and visualized in Web GUI.
- 2 : Event data is read from the ObjectServer and enriched
- Netcool/Impact reads the event data from the ObjectServer. In Netcool/Impact, the event data is enriched by information that is retrieved by Impact policies.
- 3 : Event data is visualized and managed in the Web GUI
- The Web GUI displays the application events that are in the ObjectServer. From the event lists, you can run tools that change the event data; these changes are synchronized with the data in the ObjectServer.
- Event Analytics data flow
- Event data is archived and historical event data is used to generate analytics data.
- 4 : Events are read from the ObjectServer by the Gateway for JDBC
- The Gateway for JDBC reads events from the ObjectServer.
- 5 : Event data is archived
- The Gateway for JDBC sends the event data via an HTTP interface to the Historical Event database. The figure shows an IBM® Db2® database but any supported database can be used. The gateway must be configured in reporting mode. This data flow is a prerequisite for the event analytics capability.
- 6 : Event analytics algorithms run on archived event data
- After a set of historical alerts is archived, the seasonality algorithms of the Netcool/Impact policies can generate seasonal reports. The related events function analyzes Netcool/OMNIbus historical event data to determine which events have a statistical tendency to occur together and can therefore be grouped into related event groups. Pattern functions analyze the statistically related event groups to determine whether the groups have any generic patterns that can be applied to events on other network resources.
- 7 : Analytics data is visualized and managed
- The seasonality function helps you identify and examine seasonal trends while monitoring and managing events. This capability is delivered in a Seasonal Events Report portlet in Dashboard Application Services Hub. The portlet contains existing seasonal reports, which can be used to identify the seasonal pattern of the events in the Event Viewer. You can create new seasonal reports and edit existing ones. Statistically related groups can be analyzed in the Related Events GUI. Validated event groups can be deployed as Netcool/Impact correlation rules. Patterns in the statistically related event groups can also be analyzed in the Related Events GUI. These patterns can be extracted and deployed as Netcool/Impact generalized patterns.
- Event Search data flow
- Event data is indexed in Operations Analytics - Log
Analysis and used to
display event dashboard and timelines.
- 8 : Events are read from the ObjectServer by Gateway for Message Bus
- The Gateway for Message Bus reads events from the ObjectServer.
- 9 : Event data is transferred for indexing to Operations Analytics - Log Analysis
- The Gateway for Message Bus sends the event data via an HTTP interface to the Operations Analytics - Log Analysis product where the event data is indexed. The Tivoli® Netcool/OMNIbus Insight Pack 1.3.0.0 parses the event data into a format suitable for use by Operations Analytics - Log Analysis. The diagram shows the default IDUC connection, which sends only event inserts. For event inserts and reinserts, the Accelerated Event Notification client can be deployed, which can handle greater event volumes. See On-premises scenarios for Operations Management.
- 10 : Event search data is visualized
- Event search results are visualized in Operations Analytics - Log Analysis event dashboards and timelines by performing right-click tools from event lists in Web GUI.