tencent cloud

Tencent Cloud DataBuddy

DocumentaçãoTencent Cloud DataBuddyData AnalysisData Exploration Overview

Data Exploration Overview

Baixar
Modo Foco
Tamanho da Fonte
Última atualização: 2026-10-09 16:15:01
Traduzido por IA

What Is Data Exploration

Data Exploration is a lightweight analysis entry point in DataBuddy designed for data analysts. Here, you can directly query tables in the unified Catalog using Spark SQL, instantly convert query results into charts, and save frequently used queries as favorites, without going to Workflow orchestration or Notebook development processes.
Here is a typical example: a business user wants to view the order trend for a specific product line over the last 7 days. The user only needs to create a SQL query in Data Exploration, select Catalog → Schema → Run, and the results immediately appear in the table below. Then, clicking Chart generates a line chart, which can be shared with colleagues via a link.

Relationship with SQL in Data Engineering

Data Exploration and the development workbench SQL share the same underlying capabilities: the same Spark SQL engine, the same Serverless data analytics compute resource group, the same unified Catalog, and the same query execution pipeline. The difference between the two lies in their product entry points and positioning.
Dimension
Data Exploration (5.1)
Studio SQL(3.1.3)
Target Readers
Data analysts and business analysts
Data engineers and ETL developers
Product Entry
Left navigation Data Exploration
Left navigation Development Workbench
Main Scenarios
Ad hoc queries, visual exploration, and temporary analysis
Developing data processing scripts, parameterized reuse, and scheduling preparation
Collaboration Context
Standalone SQL script files + global run history + visualization
Multi-file collaboration with Notebook / SQL / Python + Git
Relationship with Workflow
Typically not scheduled; used for idea validation.
Can be added to a workflow as an SQL task with one click.
Tip
SQL files from both entry points are interoperable in Workspace: a .sql file saved in Data Exploration can be opened and further edited in Studio, and vice versa. Both follow the same file permission rules. This means that queries saved by analysts can be directly reused by engineers and added to Workflow scheduling.

Features

Capability
Description
Ad hoc query
Write Spark SQL in the unified editor and submit it to a Serverless resource group for execution within seconds. Results are displayed as a table below the editor.
Favorite queries
Save frequently used queries to a favorites list and load them into the editor with one click next time.
Result visualization
Generate visual charts such as bar charts, line charts, and pie charts from query results with one click, without switching to the dashboard.
Export Results
Download query results as CSV / Excel / TXT for offline reporting or external delivery.
AI Assistance
Built-in Buddy "Data Analysis" mode supports natural language SQL generation, query interpretation, report generation, and attribution analysis.
Federated Query
Access external data sources (MySQL, Doris, ClickHouse, and others) through a unified Catalog, and query across sources using the same Spark SQL syntax.

Key Concepts

To understand Data Exploration, you need to be familiar with the following core concepts:
Data Exploration Space: a collection of all Data Exploration SQL files in Workspace, organized in multi-level folders. File-level ACLs are consistent with Studio SQL. It is equivalent to SQL Editor / Queries in Databricks and Data Analysis in Alibaba Cloud DataWorks.
Spark SQL Dialect: the underlying execution engine uses Spark SQL uniformly. All cross-data-source queries go through the unified Catalog and use the same syntax. For more information, see SQL IDE Basic Operations.
Interactive Analytics Serverless Resource: the execution resource used by Data Exploration. It is dynamically scaled in and out by the platform, so analysts do not need to be aware of specific instances. For more information, see Compute Resource Overview.
Unified Catalog: a three-part namespace catalog.schema.table based on Unity Catalog that hosts metadata for internal engines and external data sources. For more information, see Three-Level Namespace.
Query History and Query Favorites correspond to two types of workflows: "what has been run in the past" and "what you want to reuse." They support filtering the personal visible scope by file ACL.
For a complete list of terms, see Glossary.

Applicable Scenarios

Scenario
Applicable Method
Ad hoc business data query
Write a piece of SQL, view the result once, and no need to save it.
Recurring daily queries
Save frequently used SQL statements for instant loading each time they are opened.
Trend/proportion analysis
Generate charts directly in the result area without going to the dashboard.
Offline reporting/data delivery
Download results as CSV / Excel.
Cross-source federated query
Access external data sources through External Catalog.
Convert analysis results into reports.
Consolidate exploration results into a dashboard to form a shareable view.
Convert exploration results into production tasks.
Bind SQL files directly as Workflow SQL tasks.

References

SQL IDE Basic Operations
SQL Script Execution
SQL Task
Dashboard Overview
Three-Level Namespace
Compute Resource Overview
Anterior: Inference Monitoring and LoggingPróximo: Unified Metadata Management Architecture

Ajuda e Suporte

Esta página foi útil?

comentários

Search
    Enter keywords to search documentation