Help Center> > User Guide > Browser >Browser Summary

Browser Summary

Introduction

The Application Performance Management (APM) browser function provides intuitive performance data of the frontend from multiple dimensions, including user experience, throughput, and error, helping you quickly locate performance bottlenecks and optimize performance.

Previously, people paid more attention to server performance than frontend performance. In the Internet era when user experience is of crucial importance, user experience at the frontend still cannot be monitored even though backend services are running stably. If exceptions occur at the backend, you can quickly locate and rectify faults based on logs. However, when user experience at the frontend is not satisfactory, you may not be aware of that in uncertain application scenarios. You can only optimize the frontend based on surveys and user feedback. Such methods are inefficient or even ineffective.

To solve the problem, Huawei provides the APM browser function, enabling you to establish relationships with all users around the world and monitor their experience in real time. The Browser Summary page displays user experience in real time, helping you quickly optimize performance and achieve business success.

Browser Flowchart

Monitoring Metrics

User Experience

APM uses the Application Performance Index (Apdex) to measure user experience. APM classifies requests into satisfied, tolerating, and frustrated requests based on the response time, and quantifies user experience using certain formulas, helping you obtain basic application information.

Throughput

Throughput refers to the total number of network requests in a specified period.

TP99 Latency

TP99 latency is the minimum time required to meet requirements of 99% requests. For example, the time required for four requests is as follows: 10 ms, 100 ms, 500 ms, and 20 ms. 99% requests = 99% x 4, of which the rounding value is 4. That is, the number of 99% requests is 4. The minimum time required for the four requests is 500 ms. Therefore, TP99 latency is 500 ms.

Errors

Errors refer to the number of network request failures.

Page Loading Time

When you click a hyperlink or access a URL through the address bar, the page loading process starts, and the entire process ends until the page is rendered in your browser. APM records each time point of the page loading timeline, calculates the time consumed in each phase, and displays the result in a cumulative chart, helping you quickly locate frontend performance bottlenecks. The entire loading process is divided into four phases based on services, as shown in the following figure.

The following table lists the methods of calculating the time consumed in each phase.

Reported Field

Description

Calculation Method

Remarks

network

Time consumed by the network

(requestStart - navigationStart) + (responserEnd - responseStart)

-

dom_processing

Time consumed by DOM processing

domContentLoadedEventEnd - responseEnd

-

page_rendering

Time consumed by page rendering

domComplete-domContentLoadedEventEnd

Time consumed by first page rendering

web_application

Time consumed by backend services

responseEnd - requestStart

-

You can optimize the performance based on metrics in different phases.

Related Operations

You can click on the left of the Browser Summary page and perform the following operations.

Object

Operation

Description

Browser Summary page

Lock

Click on the left of the Browser Summary page to add, delete, or reset a widget.

Add

Delete

Reset

Widget

Delete

Click on the left of the Browser Summary page to delete or zoom in a widget.

Zoom in