Why PowerJob Makes Our Project Sleep Easy: A Hands‑On Guide

Overview

PowerJob is a Java‑based enterprise‑grade distributed task scheduling platform. It provides a web UI for visual task management, supports multiple scheduling strategies, and offers several execution modes.

Scheduling strategies : CRON, fixed frequency, fixed delay, and API‑driven triggers.

Execution modes : single‑node, broadcast, Map, and MapReduce. Map/MapReduce enable cluster‑wide computation with minimal code.

Workflow support : DAG‑based task dependencies, data passing between upstream and downstream nodes, and conditional or nested workflow nodes.

Executor compatibility : Spring Bean, built‑in/externally provided Java classes, and optional Shell, Python, HTTP, SQL processors via a single dependency.

Operational convenience : Real‑time log streaming in the console.

Lightweight dependencies : Only a relational database (MySQL, PostgreSQL, Oracle, SQL Server, etc.) is required.

High availability & performance : Lock‑free scheduling design eliminates database‑lock bottlenecks and allows horizontal scaling.

Fault tolerance : Configurable retry policies re‑execute failed tasks as long as executor nodes are available.

Comparison with Other Schedulers

Key differences highlighted by the official comparison:

PowerJob supports CRON, fixed frequency, fixed delay, and OpenAPI (same as XXL‑Job).

It provides dynamic MapReduce sharding, whereas QuartZ lacks distributed task support and XXL‑Job uses static sharding.

Online task governance and log white‑screening are supported, unlike QuartZ and SchedulerX 2.0.

Scheduling performance is lock‑free, avoiding the database‑lock bottleneck present in QuartZ and XXL‑Job.

Alerting integrates email with extensible interfaces, while other frameworks offer limited or no alert mechanisms.

Installation

PowerJob can be installed via Docker or by running the provided JAR files.

Docker : Follow the official Docker‑Compose guide at https://www.yuque.com/powerjob/guidence/docker-compose.

Jar deployment :

Clone the source code from https://github.com/PowerJob/PowerJob and build the project.

Open the project in an IDE. The powerjob-server module contains the server code, and powerjob-worker-samples provides a Spring Boot client example.

Create a database, e.g.

CREATE DATABASE IF NOT EXISTS `powerjob-daily` DEFAULT CHARSET utf8mb4;

and adjust application-daily.properties with the correct JDBC URL, username, and password.

Run the server JAR: java -jar powerjob-server.jar and verify the login page at http://localhost:7700.

Client Project Setup

Add the Maven dependency:

<dependency>
    <groupId>tech.powerjob</groupId>
    <artifactId>powerjob-worker-spring-boot-starter</artifactId>
    <version>4.3.2</version>
</dependency>

Configure the worker in application.yml (or application.properties):

powerjob:
  worker:
    enabled: true
    enable-test-mode: false
    port: 27777
    app-name: powerjob-agent-test
    server-address: 127.0.0.1:7700
    protocol: http
    max-result-length: 4096
    max-lightweight-task-num: 1024
    max-heavy-task-num: 64

Enable Spring scheduling with @EnableScheduling on the main class.

Implement a simple task by creating a bean that implements BasicProcessor:

@Component
public class SimpleJobServer implements BasicProcessor {
    @Override
    public ProcessResult process(TaskContext taskContext) throws Exception {
        String jobParams = taskContext.getJobParams();
        System.out.println("参数: " + jobParams);
        System.out.println("定时任务执行");
        return new ProcessResult(true, "定时任务执行成功");
    }
}

Start the client application; the server UI will display the new executor instance.

Creating a Task in the UI

In the server console, click “New Task”, fill in the task name, select the scheduling type (e.g., CRON), and specify the processor class name such as com.example.powerjobdemo.job.SimpleJobServer. After saving, the task appears in the list and can be executed immediately.

Task Configuration Details

Timing information : Choose among API, CRON, fixed frequency, fixed delay, workflow, or daily fixed interval.

Lifecycle : Define an effective period so the task runs only within a specific time window.

Execution configuration :

Execution type: single‑node, broadcast, Map, or MapReduce.

Map executes simple data splits across nodes; MapReduce handles complex big‑data pipelines with separate map and reduce phases.

Runtime configuration :

Node selection strategy: HEALTH_FIRST (first healthy node) or RANDOM.

Maximum instance count, concurrency limits, and execution timeout can be set per task.

For additional parameters, refer to the official documentation at https://www.yuque.com/powerjob/guidence/ysug77.