The 3D topology feature in Application Real-Time Monitoring Service (ARMS) provides a visual representation of the health and dependencies of your applications, services, and hosts. With the 3D topology, you can quickly identify services that cause failures, see which applications are affected, and pinpoint the root cause of issues for faster troubleshooting.
Quick start
View the following interactive tutorial to learn about the features of the 3D topology in Application Monitoring. For the best experience, open the tutorial in a new window.
Entry point
Log on to the ARMS console.
In the left-side navigation pane, choose . In the top menu bar, select the target region.
On the Applications page, find the target application and click 3D Topology in the Actions column.
Overview
The Overview page, which is displayed by default, shows all content from the service layer, application layer, and host layer. The upper-right corner of the page displays the number of hosts, applications, and services.
The top of the Overview page includes a timeline with a slider, a time range picker in the upper-left corner, and a search box and a summary of host, application, and service counts in the upper-right corner. The main area displays a 3D visualization of the service, application, and host layers, and the call and dependency relationships between them. You can drag the view to rotate and inspect it from any angle.
On the Overview page, you can perform the following operations:
Click the time range picker in the upper-left corner to select a specific start and end time.
Drag the slider on the timeline at the top of the page to change the time range for the current view.
Enter a keyword in the search box in the upper-right corner and press Enter to search.
Drag the topology to view the three layers from any angle.
Click any object in the view to see its related metrics in the panel on the right.
Service layer
The service layer displays information about the services that your application depends on.
Each block represents a service under an application. The more calls a service receives, the larger its block. Different colors indicate service status.
■ : Service calls are normal.
■ : The service has a high error rate.
■ : The service returns no data.
You can configure the response time threshold for services. In the left-side navigation pane, click the triangle icon next to the service layer controls to expand the settings for the response time threshold. Then, drag the slider to set the threshold.
When you click a service, the panel on the right displays the following information:
Service name
QPS: queries per second
RT(ms): response time, in milliseconds
Error: error queries per second
In the QPS, RT(ms), and Error sections, the value on the left is the average for the selected time range, while the chart on the right shows the trend over that period.
Application layer
The application layer shows the application and its upstream and downstream dependencies, including middleware. Connecting lines indicate call direction.
When you click an application, the panel on the right displays the following information:
Application name
QPS: queries per second
RT(ms): response time, in milliseconds
Error: error queries per second
In the QPS, RT(ms), and Error sections, the value on the left is the average for the selected time range, while the chart on the right shows the trend over that period.
Host layer (Docker/ECS)
The host layer displays details about the hosts for your application.
Each cube represents a host, and hosts are grouped by application. Different colors indicate host status.
■: Normal
■: Slow
■: Warning
■: Abnormal
■: Offline
When you click a host, the panel on the right displays the following information:
Host IP address and basic information:
Response time
Number of requests
Number of errors
CPU: CPU utilization
MEM: Memory usage
DISK: Disk usage
GC TIME: Garbage collection time
GC COUNT: Garbage collection count
In the CPU, MEM, and DISK sections, the value on the left is the average for the selected time range, while the chart on the right shows the trend over that period.