Segment Routing Dashboard

Integrated Performance Management

IP networks inherently implement Equal-Cost Multipath (ECMP), allowing traffic to be evenly distributed across multiple paths of equal cost. Integrated Performance Measurements (I-PM) offers a solution to accurately measure transport service performance across all ECMP paths for any given network path, ensuring comprehensive visibility and precise network monitoring.

Integrated Performance Measurement consists of three key components:

• Routing Analytics (CNC): Monitors the underlay routing across all ECMP paths.

• IP Measurements: Provides visibility through scalable active SLA measurements from any edge to any edge across all ECMP paths.

• Provider Connectivity Assurance: Correlates IP measurements and routing data to deliver proactive, actionable insights.

This article focuses on the features and insights enabled by Routing Analytics (see figure below). Documentation on IP Measurements will be available soon.

Path Query Service

The Path Query Service (PQS) enables users to query the expected latency for a specified network path, such as from a source edge node through intermediate nodes to a destination edge node (e.g., P edge, A, B, Q edge). It provides detailed hop-by-hop minimum propagation delay performance along with the overall end-to-end expected delay performance for the queried path. This granular latency insight allows users to understand the performance characteristics and whether the network can fulfil their intent.

Network Performance Grid

The path data collected by routing analytics can be aggregated into a grid based on metadata, such as from any one city edge to any other city edge. This aggregation summarizes the minimum overall delay for each path between pairs of cities, enabling macro-level network performance analysis and insights.

How It Works

1. Navigate to the Segment Routing dashboard.
   

2. Select your desired metric, metric aggregation and time interval.
   

3. Explore your entire inventory of cities.

• The blue cells represent a gradient of metric values from light blue (low metric value) to dark blue (high metric value).

• Hovering over a cell will display a line chart of the city-to-city’s metric performance over time.

• Users can select an individual cell, a column/row header, or a city on the map for more details.

City-Level Inventory and Performance

Selecting a city displays all the other cities directly connected to it.

City-to-City Segment-Level Performance

Selecting a city-to-city pair displays all paths connected between these two cities.

ECMP Path Detail Drill-Down

Selecting a path allows you to explore detailed information about that path. At the top, the change in path delay over the selected time period is displayed. Below this, the path’s ECMP tree is shown, including node names and the minimum propagation delay for each hop. At the bottom, any changes in the ECMP paths are listed in a table alongside their corresponding end-to-end propagation delays.

Clicking the View Details button will display the full details page. See Segment Routing Inventory, Management and Details for more information.

You can select two paths from the table below to compare and analyze differences in their ECMP trees across different time stamps, as well as changes in hop-by-hop delays.