What is Analytics?

Overview

NSP Analytics generates reports and dashboard views to represent network conditions. The reporting is based on various data types that the NSP collects. For example, raw data is used for evaluating current conditions, and aggregated data helps to identify patterns and trends. Analytics reports and views provide vital information about network status, traffic, application usage, and more.

This document describes NSP Analytics functions, including report generation and visualizations, and provides a reference of the available Analytics report catalogs. For information about configuring telemetry subscriptions, using baseline analytics, NSP indicators, OAM tests, metric customization, and network tests, see the NSP Data Collection and Analysis Guide.

Note: The term “NSP” in the guide refers to an NSP deployment that may or may not include an integrated NFM-P for classic management. The NFM-P is mentioned as a separate NSP component only in required contexts, for example, “NFM-P main server”, or content that describes NFM-P configuration.

Note: Some Analytics reports require a variety of statistics or data types. All data for a report must be collected using the same collection interval; otherwise, the report generation fails or produces erroneous results.

Aggregation reporting

The available data-aggregation levels, for example, Hourly, Daily, or Monthly, vary by report type and are selectable using the Granularity drop-down menu in the input prompt panel.

Aggregation-based reporting in NSP for MDM-mediated NEs and CN telemetry requires the configuration of data aggregation in the Aggregation view of Data Collection and Analysis Management. See the NSP Data Collection and Analysis Guide, and the Telemetry information on the Network Developer Portal.

For classically managed NEs, in order to create a report or dashboard based on aggregated data, you must enable and configure aggregation for the required statistics using the NFM-P Aggregation Manager, as described in How do I configure analytics aggregation?.

Telemetry reporting

Analytics includes reports based on collected telemetry data. The title of each such report includes “(NSP)” for differentiation from the equivalent legacy NFM-P report, which is based on classic mediation. The NSP reports are based on data from model-driven Nokia or multivendor NEs, and may also include classic mediation data.

Classic mediation reporting

Analytics reporting for classic mediation requires the configuration of auxiliary database collection parameters such as the following using the samconfig utility on each NFM-P main server:

See the NSP Installation and Upgrade Guide for configuration information.

Object counts in Analytics reports

You can select up to 1000 objects for a report by entering comma-delimited names at a report input prompt that has a % search in the associated text box. For example, you can specify a list of objects such as %203.0%,%MSS%,%113.71% at the Name or name pattern for NEs prompt, or in any other text box that supports % searches, including NE names, Port-LAG, port names, interface, service, SAP, and customer.

If reporting is required for more than 1000 objects, you can filter using report prompts to generate multiple reports. The Comprehensive reports described in Chapter 11, Comprehensive reports do not implement the 1000-object limit; reporting is limited only by the available system resources.

Because no 1000-object limit is imposed, a Comprehensive report may take longer to render, or may fail due to insufficient resources. In such cases, you can use the report prompts to reduce the number of selected objects. Alternatively, you can create an ad hoc report without the 1000-object limit, or contact Nokia professional services to create a custom report.

Analytics Data Dictionary

To ensure optimal reporting performance, a subset of network and service objects, for example, subscribers, services, or ports is copied from the main NSP database to the auxiliary database. Analytics synchronizes the data for all object create, update, and delete operations. To ensure that no operations are missed, the data dictionary is periodically resynchronized and is entirely rebuilt every 30 days.

NSP Analytics uses the dictionary data to populate report prompts; the data are also available for use in ad hoc reports. Deleted objects are retained for up to 30 days, so may appear in newer reports, depending upon the report design. You can use report prompts to exclude such objects.

Analytics API support

NSP Analytics functions are available for OSS using programmable APIs. For general information about developer support, see the Network Developer Portal. For API documentation, see the API documentation page.

Resource requirements

Analytics has significant data throughput and storage requirements that vary by the data type, such as telemetry, aggregated raw, or flow. An NSP deployment that includes Analytics reporting may require the following, based on the collected data types and volume:

Note: The report characteristics table in a chapter indicates which system elements are required in order to collect the required report data. For example, the Source database row in the table for an aggregation-based report indicates that an NSP auxiliary database is required, and the NSP Flow Collector required row is present in a table that describes flow statistics collection.

If you intend to enable additional reporting that is not currently available in your deployment, see the NSP Planning Guide, which includes information about the NSP Analytics resource requirements, based on the managed network size, deployment scope, and other considerations.

Network domains for Application Assurance reports

NSP Analytics supports AA reporting for residential, business, Wi-Fi, and fixed wireless access subscribers. If needed, the network domain can be selected from the Domain drop-down menu in the input prompt panel.

Note: The use of “mobile” in this document implies fixed wireless access.

The domains are defined by the value of the Subscriber Type parameter in the AA Subscriber form, the Subscriber Scale parameter in the ISA AA group policy, or both:

The following table describes the reports available for each domain.

Table 1-1: Application Assurance reports, by domain

Report type

Residential / Wi-Fi

Business

Wi-Fi (DSM)

Mobile

Business Subscribers

None

All

None

None

Devices and Domains

All

All

All

All

Firewall

None

All

None

All

Location and Congestion

None

None

All

All

Mobile Subscribers

None

None

None

All

Nodes

All

All

All

All

Performance

All

All

All

All

Subscribers

All

None

None

None

Baselining in Analytics reports

Some reports support the use of baselining. Baselining highlights deviations for the current period from the baseline data collected in the configured baseline period.

Note: Baselining in Analytics reports is different from baselines as seen in the Data Collection and Analysis Management, Baselines view.

For an Analytics report, NSP computes a baseline for data configured for reporting, for example, utilization and throughput for a port in a Port LAG Details report, or bandwidth and data for an application group in a Router Level Usage Summary report with Baseline.

A baseline configured in Data Collection and Analysis Management, Baselines provides near-real-time trend and anomaly detection from telemetry counters, for example, received-octets for the /telemetry:base/interfaces/interface telemetry type.

For more information about baselines in Data Collection and Analysis Management, see the NSP User Guide and the NSP Data Collection and Analysis Guide .

The following table describes how baseline data is presented.

Report granularity

Baseline definition

Baseline composition

Raw

Raw + Hour of Day

Baseline is composed of all data points in the Baseline Report Range for matching intervals to the hour and minute for any day of the week.

For example if May 15 14:05 is in the Report Range, and the Baseline Report Range is April 1 00:00 - April 30 11:59, the baseline minimum, maximum, and average values are based on the 30 samples from 14:05 for every day of April. The same baseline calculation is made for other data points in the Report Range.

Raw + Hour of Day + Day of Week

Baseline is composed of all data points in the Baseline Report Range for matching intervals to the hour and minute for the corresponding day of the week.

In the above example, the baseline has 4 or 5 samples depending on the day of the week (as each day occurs 4 or 5 times per month).

Hourly

Hour of Day

Baseline is composed of all data points in the Baseline Report Range for matching hours for any day of the week.

For example, if May 15 14:00 is in the Report Range, and the Baseline Report Range is April 1 00:00 - April 30 11:59, then the baseline minimum, maximum, and average values are based on the 30 samples from 14:00 for every day of April. The same baseline calculation is made for other data points in the Report Range.

Hour of Day + Day of Week

Baseline is composed of all data points in the Baseline Report Range for matching hours for the corresponding day of the week.

In the above example, the baseline has 4 or 5 samples depending on the day of the week (as each day occurs 4 or 5 times per month).

Hour of Day + Day of Month

Baseline is composed of all data points in the Baseline Report Range for matching hours for the corresponding day of the month.

In the above example, the baseline has 1 sample (note: a longer Baseline Report Range is recommended).

Daily

Day of Week

Baseline is composed of all data points in the Baseline Report Range for matching days of the week.

For example, if May 15 is in the Report Range, and the Baseline Report Range is January 1 00:00 - April 30 11:59, the baseline minimum, maximum, and average values are based on the 17 or 18 samples for every day in those 4 months.

Day of Week + Month of Year

Baseline is composed of all data points in the Baseline Report Range for matching hours for the corresponding day of the week for the matching months.

In the above example, the baseline has 4 or 5 samples for each of the months of January-April.

Day of Month

Baseline is composed of all data points in the Baseline Report Range for matching days of the month.

In the above example, the baseline has 4 samples for the 15th day of January-April.

Day of Month + Month of Year

Baseline is composed of all data points in the Baseline Report Range for the matching days and months of the year.

In the above example, the baseline is empty as May 15th does not occur (a different Baseline Report Range is required).

Monthly

Month of Year

Baseline is composed of all data points in the Baseline Report Range for matching months.

For example, if May is in the Report Range, and the Baseline Report Range is January 1 2020 - December 31 2022, the baseline minimum, maximum, and average values are based on the 3 samples for every month of May between 2020-2022.

Analytics report branding

Some reports support the inclusion of a logo. A logo image that you add is displayed at the top of each report page in an 80-pixel square. See How do I upload images for report branding? for information.