使用 Kibana 进行可视化

Kibana is the visualization and management tool for the Elastic Stack. It lets you explore your Elasticsearch data, create beautiful visualizations, and build interactive dashboards.

Think of it as the "eyes" for your Elasticsearch data, helping you understand trends, anomalies, and patterns at a glance.

Kibana organizes its features into several key applications, accessible from the left-hand navigation menu:

• Discover: Explore raw document data.
• Visualize: Create charts, graphs, and maps.
• Dashboard: Combine visualizations into a single view.
• Stack Management: Configure index patterns, users, and more.

We'll focus on Discover, Visualize, and Dashboard in this lesson.

Before you can explore data, Kibana needs to know where to look. This is done by defining an Index Pattern.

An index pattern is a string that matches one or more Elasticsearch indices. For example, logs-* would match logs-2023-01 and logs-2023-02.

You typically create index patterns under "Stack Management" > "Index Patterns".

The Discover tab is your starting point for exploring raw data. Here, you can:

• See individual documents in a table format.
• Apply search queries using KQL (Kibana Query Language) or Lucene syntax.
• Filter data by time range using the time picker.
• View field statistics and distributions to understand your data better.

It's like a powerful data explorer for your Elasticsearch indices.

Let's index a few simple documents into an index called products. Then, we can find them in Kibana's Discover tab.

Run these cURL commands in your terminal to add some data:

After indexing the products, go to Kibana's "Stack Management" > "Index Patterns".

Click "Create index pattern" and enter products. Select the @timestamp field (if present, otherwise just proceed) and create the pattern.

Now you can navigate to the Discover tab, select the products index pattern, and see your indexed documents!

Visualizations help make sense of data. Let's create a simple bar chart showing the count of documents by category.

Steps:

1. Go to Visualize > Create new visualization.
2. Choose a type, e.g., Vertical Bar.
3. Select your index pattern (e.g., products).
4. For the X-axis, select Terms aggregation on the category.keyword field.

This will show bars for "Electronics", etc., based on your data.

Kibana offers a rich set of visualization types for different data stories:

• Line Chart: Best for showing trends over time.
• Pie Chart: Represents proportions of a whole.
• Metric: Displays a single key performance indicator (e.g., total sales).
• Area Chart: Useful for cumulative totals over time.
• Data Table: Presents raw aggregated data in a tabular format.

Choose the type that best communicates your data's narrative.

A Dashboard brings together multiple visualizations into a single, interactive view. This allows you to monitor different aspects of your data simultaneously.

You can drag and drop visualizations, resize them, and arrange them to create a coherent data story. Dashboards are perfect for operational monitoring, reporting, or business intelligence.

Kibana Dashboards aren't just static images; they are highly interactive:

• Time Range: Apply a global time filter to all visualizations on the dashboard.
• Filters: Add specific filters (e.g., category: "Electronics") to narrow down the data.
• Drill-downs: Click on a bar in a chart or a slice in a pie chart to automatically filter the entire dashboard.

This interactivity helps users quickly analyze and pinpoint specific data points.

You've learned about Kibana's core functions. Let's test your understanding.

You've learned that Kibana is an indispensable tool for visualizing and managing your Elasticsearch data. We covered:

• Its main applications: Discover, Visualize, and Dashboard.
• How to set up index patterns to connect to your data.
• Exploring raw data in the Discover tab.
• Creating basic visualizations like bar charts.
• Building and interacting with dashboards for comprehensive data views.

Kibana empowers you to turn raw data into actionable insights and compelling data stories!