Handling network performance data

Use this information to understand the end to end processing of different types of performance data that is handled by Network Performance Insight® V1.3.1 along with Tivoli® Network Manager.

Before you begin

Make sure to install the required Technology Packs. See Installing the Technology Packs.

About this task

These processes cover information about configuring the devices for data collection, configuring the Network Performance Insight and Tivoli Network Manager environments to handle the data, and the data display from various visualizations that are available in Network Performance Insight Dashboards.

These tasks are required in the following scenarios only:

Network Performance Insight to collect NetFlow data and performance data by integrating with Tivoli Network Manager.
Network Performance Insight to collect NetFlow data and performance data by integrating with Cacti.
Network Performance Insight to collect NetFlow data and extend to performance data later.

The following standard SNMP metrics are required:

cpuBusy
ifOutDiscards
ifInDiscards
snmpOutBandwidth
snmpInBandwidth
ifInErrors
ifOutErrors

Procedure

Handling standard SNMP metrics.

Configure Storm Spout in Tivoli Network Manager to enable the integration with Network Performance Insight.

See Enabling the integration with Network Performance Insight
Configure the SNMP metrics through Network Polling on Tivoli Network Manager system.
For more information, see:
- Creating fully featured poll policies
- Creating new poll definitions
  Note: Make sure to select all the required device classes that are applicable to the devices in your network from the Classes tab in the Poll Definition Editor.
The polled metrics are then stored in Network Performance Insight database.
Configure the thresholds for anomaly detection for these metrics on Tivoli Network Manager system.
For more information, see, Defining performance thresholds for anomaly detection.
Configure the network discoveries on Tivoli Network Manager system for all the supported IP SLA probe operations.

For more information, see Activating SLA agent.

Note: Polling for IP SLA metrics is performed by Network Performance Insight. For more information, see IP SLA data flow.
Configure network discoveries and polling on Network Performance Insight system for Performance Metric OOTB Device Support metrics and poll the network. Configure the thresholds for these metrics on Network Performance Insight system.
For more information, see:
Optional: Entity metric fields from Cacti

These steps are required only for the scenario where Network Performance Insight is used to collect NetFlow and performance data from Cacti.
1. Collect the performance metrics from imported data templates in Cacti. See Importing performance data templates.
2. Configure Cacti servers from System Configurations on Dashboard Application Services Hub. See Configuring Cacti servers.
3. Configure the entity thresholds for the performance metrics that are collected from Cacti. See Configuring Entity thresholds.
4. Configure ui.default-entity-scope setting on Ambari for visualization of data that is collected from Cacti. See Configuring the default entity scope for data visualization on Ambari.