Help Center/ Application Performance Management/ User Guide/ Topology
Updated on 2022-04-28 GMT+08:00
View PDF
Share

Topology

A topology graphically displays call and dependency relationships between applications. In a topology, each circle represents a service, each segment in a circle represents an instance, and each arrow represents a call relationship. APM supports calls between applications. The topology can display service call relationships across applications. When a circle represents an application, right-click the circle and choose View Application to go to the topology page.

Different colors on the circle represent different health statuses of instances. Colors are determined by Application Performance Index (Apdex) values. If an Apdex value is closer to 1, the corresponding application is healthier.

Topology Page

  1. Table 1 provides topology description.
    Table 1 Topology description

    Color

    Instance

    Call

    Green

    0.75 ≤ Apdex ≤ 1

    The instance responds quickly when it is called.

    0.75 ≤ Apdex ≤ 1

    Quick response.

    Yellow

    0.3 ≤ Apdex < 0.75

    The instance responds slowly when it is called.

    0.3 ≤ Apdex < 0.75

    Slow response.

    Red

    0 ≤ Apdex < 0.3

    The instance responds very slowly when it is called.

    0 ≤ Apdex < 0.3

    Very slow response.

    Gray

    The instance is not called.

    N/A

    Black

    The instance is deleted.

    N/A
  2. On the right of the topology page, set a time range to view the following topology details of an application:
    • Transaction Apdex
    • Top 5 services ranked by errors and latency
    • Top 5 transactions ranked by errors and latency
    • Top 5 SQL statements ranked by response time, calls, and errors
  3. In the topology, click a circle (a service) to view metric data, including Service Level Agreement (SLA) metrics, basic service metrics, and transaction details.
  4. In the topology, click a segment (an instance) in a circle to view metric data, including basic instance metrics, JVM metrics, node metrics, and transaction details.

Locating Faults Using the Topology

The following describes how to locate an instance with a slow response:

  1. Log in to the APM console.
  2. In the navigation pane, choose Topology.
  1. In the upper right corner of the topology page, set a time range during which a problem occurs.
  2. Check the instance with a long execution time (that is, the instance highlighted in red) in the topology.

  3. (Optional) For the service containing multiple instances, right-click the service and choose Expand from the shortcut menu to view call relationships between instances to preliminarily identify the abnormal instance.

  4. Choose Find Call-Chain from the shortcut menu. On the page that is displayed, further locate the fault based on call duration and other parameters.

Configuring Transaction Apdex Threshold

The response time of different transactions is different. APM enables you to configure different Apdex thresholds for different transactions. For example, if a login takes more than 50 ms, the response is slow. If a query transaction takes more than 10 ms, the response is slow. In this case, you need to set different Apdex thresholds for the login and query transactions.

  1. In the topology page, move the mouse cursor over a circle, right-click it, and click Edit Threshold.
  2. Modify the transaction Apdex threshold and click Apply.