🪐 Uniflow

A high-performance, extremely flexible, and easily extensible universal workflow engine.

📝 Overview

Uniflow efficiently handles a wide range of tasks from short-term jobs to long-term processes. It allows for declarative definition and dynamic modification of data flows, leveraging built-in extension capabilities to easily implement complex workflows. Moreover, it offers flexibility to expand functionality by adding new nodes or removing existing ones as needed.

Provide a personalized experience through your service and consistently expand its capabilities.

🎯 Core Values

• Performance: Achieve optimal throughput and minimal latency across various environments.
• Flexibility: Dynamically modify and adjust workflows in real-time.
• Extensibility: Extend system functionality through new components.

🚀 Quick Start

🛠️ Build and Install

Go 1.22 or higher is required. Follow these steps to build the source code:

git clone https://github.com/siyul-park/uniflow

cd uniflow

make init
make build

After building, the executable will be located in the dist directory.

⚡ Running an Example

Let's run a basic HTTP request handler example provided in ping.yaml:

- kind: listener
  name: listener
  protocol: http
  port: 8000
  ports:
    out:
      - name: router
        port: in

- kind: router
  name: router
  routes:
    - method: GET
      path: /ping
      port: out[0]
  ports:
    out[0]:
      - name: pong
        port: in

- kind: snippet
  name: pong
  language: text
  code: pong

To execute the workflow, use the following command:

uniflow start --from-nodes example/ping.yaml

To verify it's working, call the HTTP endpoint:

curl localhost:8000/ping
pong#

⚙️ Configuration

You can configure settings through the .uniflow.toml file or system environment variables.

TOML Key	Environment Variable Key	Example
database.url	DATABASE.URL	mem:// or mongodb://
database.name	DATABASE.NAME	-
collection.nodes	COLLECTION.NODES	nodes
collection.secrets	COLLECTION.SECRETS	secrets

📊 Benchmark

Here are the benchmark results conducted on a Contabo VPS S SSD (4 cores, 8GB) environment using the Apache HTTP server benchmarking tool. The benchmark measures the performance of the ping.yaml workflow, consisting of listener, router, and snippet nodes.

ab -n 102400 -c 1024 http://127.0.0.1:8000/ping

This is ApacheBench, Version 2.3 <$Revision: 1879490 $>
Benchmarking 127.0.0.1 (be patient)
Server Hostname:        127.0.0.1
Server Port:            8000
Document Path:          /ping
Document Length:        4 bytes
Concurrency Level:      1024
Time taken for tests:   13.760 seconds
Complete requests:      102400
Failed requests:        0
Total transferred:      12288000 bytes
HTML transferred:       409600 bytes
Requests per second:    7441.92 [#/sec] (mean)
Time per request:       137.599 [ms] (mean)
Time per request:       0.134 [ms] (mean, across all concurrent requests)
Transfer rate:          872.10 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:        0    1   3.1      0      34
Processing:     0  136  58.2    137     550
Waiting:        0  135  58.2    137     550
Total:          0  137  58.0    139     553

Percentage of the requests served within a certain time (ms)
  50%    139
  66%    162
  75%    174
  80%    181
  90%    202
  95%    223
  98%    264
  99%    295
 100%    553 (longest request)

📚 Learn More

• Getting Started: Introduces CLI installation and workflow management.
• Key Concepts: Explains the fundamental concepts such as nodes, connections, ports, and packets.
• Architecture: Details the process of loading node specifications and executing workflows.
• User Extensions: Guides on adding new features and integrating with existing services.

🌐 Community and Support

For questions or support regarding the project, use the following channels:

• Discussion Forum: Share questions and feedback.
• Issue Tracker: Submit bug reports and feature requests.

📜 License

This project is distributed under the MIT License. You are free to use, modify, and distribute it under the terms of the license.