DocsICP FilingConsole
官方文档
首页

Data Analysis

更新时间：
复制 MD 格式
产品详情
我的收藏

Data Analysis is the compute engine of EventHouse. It provides high-performance SQL queries, stream processing, and federated queries for real-time aggregation, analysis, and visualization of event data in EventStore.

Scenarios

  • Real-time Anomaly Detection: Time window functions calculate real-time metrics such as transaction success rate and latency distribution. If an analysis result triggers a threshold, such as more than 100 failed orders within 1 minute, the rules engine automatically triggers an alert.

  • Cross-Source Association Analysis: Federated queries join real-time event data in EventHouse with user dimension tables in RDS and log files in OSS without migrating the data to a single storage location.

  • Event Full-Text Index: The SEARCH syntax performs a full-text search on event content to quickly locate event records that contain specific keywords.

Prerequisites

  • The EventBridge service is enabled.

  • You have created a catalog, a namespace, and an event table in the data catalog. For more information, see Data Catalog.

Access the Data Analysis Page

  1. Log on to the EventBridge console.

  2. In the navigation pane, choose EventStore > Data Analysis.

The page has three areas:

Area

Location

Description

Catalog Panel

Left

Displays the data catalog tree, with levels: Catalog > Namespace > Event Table > Columns. Supports search, refresh, and collapse.

SQL Editor

Top right

The SQL query editor, including the query range selector, time range selector, and Run button.

Output Panel

Bottom right

Displays query results in Table or Visualization mode.

Execute SQL Queries

  1. In the Query Range Selector area above the SQL editor, select the target catalog and namespace.

    Run button is enabled only after you select a query range.

  2. In the Time Range Selector, select a query time range.

    The following preset and custom time ranges are available:

    Category

    Available Range

    Minutes

    Last 5 minutes, Last 10 minutes, Last 15 minutes

    Hours

    Last 1 hour, Last 6 hours, Last 24 hours

    Day

    Half day, Today, Last 3 days

    Week

    This week, Last 2 weeks, Last 3 weeks

    Month

    This month, Last 2 months, Last 3 months

    Custom

    Specify start and end times using the date picker.

    The default time range is Last 3 days.

  3. Enter a query statement in the SQL editor.

    The SQL editor is based on Monaco Editor and supports syntax highlighting and line number display. For more information about the query syntax, see Query syntax.

  4. Click the Run button to execute the query.

    The Run button supports two execution modes:

    Mode

    Description

    Full Query

    Returns all matching query results.

    LIMIT 1000

    Automatically adds LIMIT 1000 to the end of the query statement, returning only the first 1000 results.

    Click the drop-down arrow to the right of the Run button to switch between execution modes.

View Query Results

After the query is executed, results appear in the Output Panel. The Output Panel has two tabs: Table and Visualization.

Table Tab

The Table tab displays query results and supports the following features:

Feature

Description

Number of Events

Displays the total number of query results. Supports collapse.

Display Mode

Supports two modes: Table (default) and Raw.

Filter Columns

Show or hide specific columns.

Paging

Set the number of items per page (10 by default) and navigate through results by using page numbers.

Visualization Tab

The Visualization tab displays query results as charts.

  1. Click the Visualization tab, and then click Add Chart.

  2. In the Edit Chart dialog box, select a chart type and configure the chart parameters.

    The following chart types are supported:

    Chart Type

    Description

    Line Chart

    Suitable for displaying change trends of data over time. Default chart type.

    Column Chart

    Suitable for comparing data volume across different categories.

    Pie Chart

    Suitable for displaying the proportion of each part to the whole.

    The chart editor provides the following configuration items. A line chart is used as an example:

    Configuration Category

    Configuration Item

    Panel Configuration

    Panel title, panel description, display background border.

    Graphic Configuration

    Line style (line width, line style, connection method), fill style, point style, data labels, stacking, gridline.

    Legend Configuration

    Display legend, legend mode (list/table), legend position (bottom/top), legend sorting.

    Tooltip Configuration

    Sorting, display title, decimal place.

    X-Axis Configuration

    Display X-axis, X-axis title, number of ticks, format.

    Y-Axis Configuration

    Display Y-axis, unit, number of ticks, Y-axis range.

  3. After you complete the configuration, click OK to save the chart.

To add multiple Visualization tabs, click the + button in the Output Panel.

Multi-Tab Analysis

Data Analysis supports multiple analysis tabs at the same time. Each tab independently maintains its own query range, time range, SQL content, and query results.

  • Click the + button in the SQL editor area to create a new analysis tab.

  • Click the Close button on a tab to close it.

Catalog Panel Operations

The Catalog panel displays the data catalog hierarchy: Catalog > Namespace > Event Table > Columns.

Operation

Description

Search

Enter keywords in the search box to filter the directory tree.

Refresh

Click the Refresh button to update the directory tree.

Expand Table

Expand the Event Table node to view all column names and data types.

View Table Information

Hover over a table name to view its description; supports copying the table name.

Show/Hide

Toggle the Catalog Panel by using the toolbar button.

Query Syntax

Data Analysis supports SQL queries compatible with PostgreSQL syntax and provides extended full-text search capabilities. For more information, see the PostgreSQL Official Manual.

SQL Query

Data Analysis supports standard PostgreSQL syntax, including clauses such as SELECT, WHERE, GROUP BY, ORDER BY, JOIN, and LIMIT.

Example: Query the event data in a specified table.

SELECT order_id, amount, status
FROM "order-events"
LIMIT 10
Note

  • Table names must be enclosed in double quotation marks.

  • A SELECT * statement is automatically converted to a SELECT statement that specifies column names. To improve query efficiency, specify only the required columns in your query.

  • All queries are automatically limited to the selected catalog and namespace.

  • All queries require a specified time range (__time__). You can set this time range using the time range selector in the console and do not need to manually specify it in the SQL statement.

  • When you join external tables in a federated query, you do not need to specify __time__ for the external tables.

Full-Text Index

The SEARCH keyword performs a full-text search on event content.

Syntax:

SELECT * FROM "table_name" SEARCH 'keyword'

Example: Retrieve event records that contain "error":

SELECT * FROM "order-events" SEARCH 'error'
Note: The search content after the SEARCH keyword must be enclosed in single quotation marks.

Query Syntax Summary

Query Type

Syntax Description

Example

Full-text index

Use the SEARCH keyword to retrieve event content.

SELECT * FROM "table" SEARCH 'keyword'

Standard SQL

Compatible with PostgreSQL SQL syntax.

SELECT col1, col2 FROM "table" WHERE col1 > 100

Federated Query

Associate internal tables with external data sources using JOIN.

SELECT * FROM "events" JOIN "external_table" ON ...
Previous：Data catalogNext：Data Intelligence Luma