A Practical Guide to Distributed Scheduled Tasks and Integrating XXL‑JOB in the Austin Project

01. How to Implement Simple Scheduling?

When I first learned Java, I used the JDK's native Timer class to demonstrate scheduling. Later, after studying concurrency, I discovered the more powerful ScheduledExecutorService interface, which runs on a thread pool unlike the single‑threaded Timer . Eventually I learned about Quartz , which supports cron expressions and is commonly used for production‑grade scheduling.

The main advantage of ScheduledExecutorService over Timer is that it uses a thread pool, allowing more efficient resource utilization.

When I later needed a real project, I turned to Quartz and eventually to Spring's @Schedule annotation, which also supports cron expressions with minimal configuration.

02. Internship & Work – Real‑World Uses of Scheduled Tasks

In production I frequently use scheduled tasks for:

Dynamic creation of push‑notification jobs for operations.

Ad settlement jobs that scan tables and update statuses.

Daily data‑record updates.

Periodic scans to achieve eventual consistency without distributed transactions.

Our in‑house distributed scheduling framework extends Quartz with failover, sharding, and other mechanisms.

03. Why Distributed Scheduled Tasks?

Single‑machine solutions like Timer , ScheduledExecutorService , and Spring @Schedule work only in standalone deployments. In clustered production environments we need a task to run on exactly one node, which requires a distributed approach.

Quartz offers a cluster mode; alternatively, developers use database row locks or Redis distributed locks to ensure single execution.

04. Basics of Distributed Scheduling

Quartz abstracts three roles: Scheduler, Executor, and Job. Users typically add a client library, annotate or implement an interface, and provide their own job logic.

05. Choosing a Distributed Scheduling Framework

Popular open‑source options include XXL‑JOB, Elastic‑Job, LTS, SchedulerX, Saturn, PowerJob, and many others built on top of Quartz . For the Austin project we selected xxl‑job because of its maturity, stability, and active community.

06. Why Austin Needs a Distributed Scheduler

The Austin admin console manages message templates, many of which require timed pushes. Therefore the scheduler must support dynamic creation, pausing, deletion, and editing of tasks directly from the UI.

07. Integrating XXL‑JOB into Austin

Steps to integrate:

Add the xxl‑job‑core Maven dependency.

Execute the provided /xxl-job/doc/db/tables_xxl_job.sql script in MySQL.

Download the xxl‑job source, configure the xxl‑job‑admin database, and start the admin service.

Add xxl‑job configuration to the Austin project.

Annotate job methods with @XxlJob to define the task logic.

XXL‑JOB follows the “centralized” model: the scheduler (admin) and executors are separate, communicating via HTTP or RPC.

08. Summary

The article covered what scheduled tasks are, why they are essential, Java‑level implementations (Timer, ScheduledExecutorService, Quartz, Spring @Schedule), fundamentals of distributed scheduling, framework selection, and a concrete integration of xxl‑job into the Austin project. Readers are encouraged to explore the open‑source code and try other frameworks.

Next time I will discuss how to handle the user‑group file generated by a triggered task and design the message‑push workflow.

austin project Gitee link: https://gitee.com/zhongfucheng/austin

austin project GitHub link: https://github.com/ZhongFuCheng3y/austin