Get started with Operational Intelligence

Complete the following setup and initial configuration steps to start using Operational Intelligence.

About this task

Raw metric data, collected by data sources (such as SCOM) needs to reach the Operational Intelligence MID Server, where Operational Intelligence can process and analyze it. You can set up Operational Intelligence so that raw data is either pulled by the Operational Intelligence MID Server from the external data source, or pushed from the external data source to the Operational Intelligence MID Server.

Using either method (pulling or pushing) requires the configuration of the ITOA MetricExtension MID Server extension. The main roles of this extension are to normalize the raw data, and to detect any anomalies. The normalized data and anomalies (if above the specified threshold), are then transmitted to the instance.

Some of the steps below for setting up Operational Intelligence, must be completed regardless of the method used. Other steps depend on the method used for transmitting data between the data source and the MID Server:

Pulling

This method requires using a connector that is configured to pull data from the specific data source that is collecting the raw data. You can use a pre-defined connector such as SCOM (which was previously defined for pulling events from SCOM), Azure, or AWS, or configure a new custom connector specifically for the external source that is used in your environment.

Pushing

This method uses web service APIs and is based on client-side tools that push the raw data from the external source to the MID Server. This method requires that the ITOA MetricExtension extension is configured with the Enable REST Listener option enabled.

You also need to configure the Web Server extension, which starts a Web Server on the MID Server. Request handlers are added on the Web Server that listen to any incoming metric data that was pushed from the external source. The configuration of the Web Server extension includes two important settings:
  • Authentication type, which can be set to the more advanced option - Keybased.
  • Secure Connection, which lets you choose whether incoming and outgoing data is secured when transmitted. If you choose the more advanced secured option, it requires that you obtain a certificate from a well known certificate authority, and then provide the Keystore Certificate Alias and the Keystore Password.

To set up Operational Intelligence you can complete the following procedure. Alternatively, you can follow a prescribed step-by-step guidance. Navigate to Guided Setup > ITOM Guided Setup, and click Continue. Then scroll to Operational Intelligence, and click Continue.

Procedure

  1. Activate the Operational Intelligence plugin.
  2. Configure a MID Server and a MID Server distributed cluster for Operational intelligence: Create a MID Server cluster with the Distributed type. Then add to that cluster at least one MID Server with the required hardware, software, and other configurations, and with the ITOA Metric capability from the Operational Intelligence plugin. It is possible to use a dedicated MID Server for both Root Cause Analysis (RCA) and Operational Intelligence. If the default SCOM connector is used, then the same MID Server that is used to collect events, should be used for Operational Intelligence.
  3. Configure the Operational Intelligence extension.
  4. To configure for pulling:
    1. Create a connector definition (For a SCOM connector - this is already defined for pulling events from SCOM).
    2. Configure a connector instance (For a SCOM connector - see Configure the SCOM connector instance).
    3. Configure an Azure connector to pull metric data from Microsoft Azure.
    4. Configure an AWS connector to pull metric data from Amazon Web Services (AWS)
  5. To configure for pushing:
    1. Configure the MID Web Server extension.
    2. In the ITOA MetricExtension extension, ensure that the Enable REST Listener option is enabled, and that the MID Web Server Extension is set.
    3. Create the client scripts that push the raw data from the external source to the MID Server.
  6. Create or configure metric type registration.
  7. Choose and configure metrics to monitor.
  8. Create event rules to map raw metric data to specific CI types.