Configs2026年4月10日·1 分钟阅读

Redash — Open Source Data Visualization & Dashboard Tool

Redash connects to any data source, lets you query with SQL, visualize results, and build shareable dashboards. The SQL-first open-source BI tool for data teams.

AI
AI Open Source · Community
快速使用

先拿来用,再决定要不要深挖

这里应该同时让用户和 Agent 知道第一步该复制什么、安装什么、落到哪里。

git clone https://github.com/getredash/setup.git redash-setup
cd redash-setup
docker compose up -d

Open http://localhost:5000 — create your admin account and connect your first data source.

介绍

Redash is an open-source data visualization and dashboarding tool designed for data analysts and engineers. It connects to virtually any data source, provides a powerful SQL editor with autocomplete, and lets you build interactive dashboards that can be shared across your organization.

With 28.3K+ GitHub stars and BSD-2-Clause license, Redash is the go-to SQL-first BI tool for teams that prefer writing queries over drag-and-drop builders.

What Redash Does

  • SQL Editor: Full-featured SQL editor with schema browser, autocomplete, and query history
  • 50+ Data Sources: PostgreSQL, MySQL, BigQuery, Snowflake, Redshift, ClickHouse, MongoDB, APIs, and more
  • Visualizations: Charts, tables, maps, pivot tables, word clouds, and more from query results
  • Dashboards: Compose visualizations into interactive dashboards with filters and parameters
  • Alerts: Automated alerts when query results match conditions (email, Slack, webhook)
  • Scheduled Queries: Auto-refresh queries on a schedule (hourly, daily, weekly)
  • Sharing: Share queries and dashboards with team via links or embedding
  • API: Full REST API for programmatic access to queries and results
  • Parameters: Parameterized queries for dynamic, interactive dashboards

Architecture

┌──────────────┐     ┌──────────────┐     ┌──────────────┐
│  Browser     │────▶│  Redash      │────▶│  Your Data   │
│  Dashboard   │     │  Server      │     │  PostgreSQL  │
└──────────────┘     │  (Python/    │     │  MySQL       │
                     │   Flask)     │     │  BigQuery    │
                     └──────┬───────┘     │  Snowflake   │
                            │             │  ClickHouse  │
                     ┌──────┴───────┐     │  50+ more    │
                     │  Redis +     │     └──────────────┘
                     │  PostgreSQL  │
                     │  (Metadata)  │
                     └──────────────┘

Self-Hosting

Docker Compose

services:
  redash:
    image: redash/redash:latest
    command: server
    ports:
      - "5000:5000"
    environment:
      REDASH_DATABASE_URL: postgresql://redash:redash@postgres/redash
      REDASH_REDIS_URL: redis://redis:6379/0
      REDASH_SECRET_KEY: your-secret-key
    depends_on:
      - postgres
      - redis

  worker:
    image: redash/redash:latest
    command: worker
    environment:
      REDASH_DATABASE_URL: postgresql://redash:redash@postgres/redash
      REDASH_REDIS_URL: redis://redis:6379/0

  scheduler:
    image: redash/redash:latest
    command: scheduler
    environment:
      REDASH_DATABASE_URL: postgresql://redash:redash@postgres/redash
      REDASH_REDIS_URL: redis://redis:6379/0

  postgres:
    image: postgres:16-alpine
    environment:
      POSTGRES_USER: redash
      POSTGRES_PASSWORD: redash
      POSTGRES_DB: redash
    volumes:
      - pg-data:/var/lib/postgresql/data

  redis:
    image: redis:7-alpine

volumes:
  pg-data:

Key Features

Parameterized Queries

-- Parameters appear as {{param_name}} in queries
SELECT
  date_trunc('{{period}}', created_at) as period,
  COUNT(*) as orders,
  SUM(total) as revenue
FROM orders
WHERE created_at BETWEEN '{{start_date}}' AND '{{end_date}}'
  AND status = '{{status}}'
GROUP BY 1
ORDER BY 1

-- Parameters become dropdown/date picker controls in the dashboard
-- period: day, week, month (dropdown)
-- start_date, end_date: date pickers
-- status: completed, pending, cancelled (dropdown from query)

Visualization Types

Type Best For
Line/Area Chart Time series trends
Bar Chart Category comparisons
Pie/Donut Proportions
Scatter Correlations
Pivot Table Multi-dimensional analysis
Map (Choropleth) Geographic data
Counter Single KPI number
Funnel Conversion analysis
Word Cloud Text frequency
Cohort Retention analysis

Scheduled Queries & Alerts

Query: "Daily Active Users"
Schedule: Every day at 8am
Alert: When result < 1000
  → Send to Slack #metrics
  → Email to team@company.com

Query Snippets

-- Save reusable SQL snippets
-- Snippet: "date_filter"
WHERE created_at BETWEEN '{{start_date}}' AND '{{end_date}}'

-- Snippet: "user_join"
LEFT JOIN users u ON u.id = t.user_id

-- Use with trigger keyword in SQL editor

Redash vs Alternatives

Feature Redash Metabase Grafana Superset
Query approach SQL-first Visual + SQL PromQL + SQL SQL + visual
Data sources 50+ 20+ 100+ 30+
Ease of use Medium (SQL) Easy (no-code) Medium Medium
Best for Data/SQL teams Business users DevOps/SRE Data teams
Alerts Yes Yes Yes Yes
Embedding Basic Advanced Yes Yes
Community Large Very large Very large Large

常见问题

Q: Redash 和 Metabase 怎么选? A: 如果你的团队习惯写 SQL,Redash 更自然。如果团队中有非技术人员需要自主探索数据,Metabase 的可视化查询构建器更友好。两者都可以免费自托管。

Q: Redash 还在积极维护吗? A: Redash 被 Databricks 收购后,官方开发节奏放缓。但社区仍在活跃维护(合并 PR 和修复 bug)。对于新项目,也可以考虑 Metabase 或 Apache Superset。

Q: 可以可视化 API 数据吗? A: 可以。Redash 支持 JSON API 和 Python 脚本作为数据源。你可以写 Python 代码调用任何 API 并返回结果用于可视化。

来源与致谢

讨论

登录后参与讨论。
还没有评论,来写第一条吧。

相关资产