garomonegro / numaflow

Kubernetes-native platform to run massively parallel data/streaming jobs

Numaflow

Summary

Numaflow is a Kubernetes-native tool for running massively parallel stream processing. A Numaflow Pipeline is implemented as a Kubernetes custom resource and consists of one or more source, data processing, and sink vertices.

Numaflow installs in a few minutes and is easier and cheaper to use for simple data processing applications than a full-featured stream processing platforms.

Use Cases

• Real-time data analytics applications.
• Event driven applications such as anomaly detection, monitoring, and alerting.
• Streaming applications such as data instrumentation and data movement.
• Workflows running in a streaming manner.

Key Features

• Kubernetes-native: If you know Kubernetes, you already know how to use Numaflow.
• Language agnostic: Use your favorite programming language.
• Exactly-Once semantics: No input element is duplicated or lost even as pods are rescheduled or restarted.
• Auto-scaling with back-pressure: Each vertex automatically scales from zero to whatever is needed.

Data Integrity Guarantees:

• Minimally provide at-least-once semantics
• Provide exactly-once semantics for unbounded and near real-time data sources
• Preserving order is not required

Roadmap

• User-defined Source (v0.11)

Demo

Resources

• QUICK_START
• EXAMPLES
• DEVELOPMENT
• CONTRIBUTING