realtime-streaming-event-driven
Specializes in Real-Time Streaming Architectures, Batch-Stream Unification, and Streaming Databases based on Streaming Databases (Hubert Dulay) and Building Real-Time Analytics Systems (Mark Needham). Covers Apache Kafka, Apache Pinot, Apache Flink, ClickHouse, Change Data Capture (CDC via Debezium), and real-time OLAP queries with sub-second latency.
来源信息
- 仓库
- dandgabr/Coacus
- 最近来源活动
- 2026年9月28日 14:03
- 检测到的 SKILL.md 语言
- 英语
- 星标
- 4
- 分支
- 3
安装方式
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
检查来源文件
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
文件资源管理器
5 个文件正在显示 SKILL.md
SKILL.md
来源说明 · 只读预览- name
- realtime-streaming-event-driven
- description
- Specializes in Real-Time Streaming Architectures, Batch-Stream Unification, and Streaming Databases based on Streaming Databases (Hubert Dulay) and Building Real-Time Analytics Systems (Mark Needham). Covers Apache Kafka, Apache Pinot, Apache Flink, ClickHouse, Change Data Capture (CDC via Debezium), and real-time OLAP queries with sub-second latency.
# Real-Time Streaming Architectures and Streaming Databases
This skill establishes patterns for designing and operating systems oriented toward continuous event streams, unified batch-stream processing, and real-time OLAP analytics with millisecond-scale latencies.
---
## ⚡ 1. Lambda vs Kappa vs Unification in a Streaming DB
```
┌─────────────────────────────────────────────────────────────┐
│ Operational Sources (OLTP: Postgres, MySQL, APIs, IoT) │
└──────────────────────────────┬──────────────────────────────┘
│ CDC (Debezium) / Event Producer
┌──────────────────────────────▼──────────────────────────────┐
│ Distributed Event Log (Apache Kafka / Apache Pulsar) │
└──────────────────────────────┬──────────────────────────────┘
│
┌──────────────────┴──────────────────┐
│ │
┌───────────▼───────────┐ ┌───────────▼───────────┐
│ Stateful Processor │ │ Streaming OLAP DB │
│ (Apache Flink) │ │ (Apache Pinot / │
│ (Windows, Aggregation)│ │ ClickHouse) │
└───────────────────────┘ └───────────┬───────────┘
│
┌───────────▼───────────┐
│ Dashboards & APIs │
│ (< 50ms Query Latency)│
└───────────────────────┘
```
---
## 📊 2. Change Data Capture (CDC) Patterns with Debezium
- **Log-Based CDC**: Directly reads the database's transaction logs (WAL in PostgreSQL, Binlog in MySQL) without the overhead of polling `SELECT` queries.
- **Outbox Pattern**: Atomic write to the `outbox` table within the same relational database transaction to guarantee that domain events are never lost during network failures.
在 GitHub 查看