Java实现的并发任务处理实例
2024-09-02 00:19:09
本文实例讲述了Java实现的并发任务处理方法。分享给大家供大家参考，具体如下：
public void init() {
super.init();
this.ioThreadPool = new ThreadPoolExecutor(50, 50, Long.MAX_VALUE, TimeUnit.SECONDS, new java.util.concurrent.LinkedTransferQueue<Runnable>(), new ThreadFactory() {
AtomicLong id = new AtomicLong();
@Override
public Thread newThread(final Runnable r) {
Thread t = new Thread(new Runnable() {
@Override
public void run() {
try {
r.run();
} catch (RuntimeException e) {
logger.error("执行IO异常", e);
throw e;
}
}
});
t.setDaemon(true);
t.setName("FootballService-IO-" + id.getAndIncrement());
return t;
}
});
}
//从fdate到tdate， 结果： 日期 30个区的创建角色总和
public Map<String, Long> countCreateRole(String fdate, String tdate, final String channel, Map<String, Object> gameConfig) throws Exception {
// 只读数据不需要处理并发
final Map<String, String> param = new HashMap<String, String>();
param.put("fdate", fdate);
param.put("tdate", tdate);
param.put("channel", channel);
final Map<String, Long> date_count = new HashMap<String, Long>();
final List<String> zones = (List<String>) gameConfig.get("areas");
Set<String> dateSet = JdbcTool.getDateRangeByDay(fdate, tdate, "yyyy-MM-dd");
List<Future<Void>> tasks = new ArrayList<>(zones.size());
for (String date : dateSet) {
final String _date = date;
tasks.add(publicThread.submit(new Callable<Void>() {
@Override
public Void call() {
final AtomicLong count = new AtomicLong();
List<Future<Void>> subTasks = new ArrayList<>(zones.size());
for (String _zone : zones) {
final String zone = _zone;
subTasks.add(ioThreadPool.submit(new Callable<Void>() {
@Override
public Void call() throws Exception {
JdbcTemplate _jdbcTemplate = dataSourceManager.getJdbcTemplate(zone);
String database = dataSourceManager.getDatabase(zone);
String _count = mget(CacheConstant.RZRoleCreateCount, zone + "#" + _date + "#" + channel + "#");
if (_count == null) {
StringBuilder sb = new StringBuilder();
sb.append("SELECT count(roleId) as count ");
sb.append("from " + database + "_log.role ");
sb.append("WHERE DATE(createTime)='" + _date + "' ");
if (param.get("channel") != null) {
sb.append(" AND channelId = '" + channel + "' ");
}
long queryForLong = _jdbcTemplate.queryForLong(sb.toString());
count.addAndGet(queryForLong);
mput(CacheConstant.RZRoleCreateCount, zone + "#" + _date + "#" + channel + "#", queryForLong + "");
} else {
count.addAndGet(Long.valueOf(_count));
}
return null;
}
}));
}
for (Future<Void> task : subTasks) {
try {
task.get();
} catch (Exception e) {
throw new RuntimeException(e);
}
}
synchronized (date_count) {
date_count.put(_date, count.get());
}
return null;
}
}));
}
for (Future<Void> task : tasks) {
task.get();
}
return date_count;
}
@PreDestroy
public void destroy() {
this.ioThreadPool.shutdownNow();
}
最新文章
- category中重写方法?
- mix_alternates_for_parent: TRUE
- bash脚本中的普通数组和关联数组
- 判断i在字符串中出现的次数（2016.1.12P141-1）
- 收藏的博客--PHP
- js中的referrer返回上一页使用介绍
- Maven详解 之 聚合与继承
- System Operations on AWS - Lab 1W - Creating EC2 (Windows)
- C# ashx生成的验证码
- SQL拼接方法
- compass scss blueprint
- 01——Solr学习之全文检索服务系统的基础认识
- （矩阵快速幂）HDU1575 Tr A
- mysql数据库备份与导入
- JavaScript：原生JS实现Facebook实时消息抓捕
- lamp安装总结
- classpath路径配置
- 大数据入门到精通4--spark的rdd的map使用方式
- anaconda使用以及创建python3.7+pytorch1.0虚拟环境以及Jupyter notebook初级使用
- [分布式系统学习] 6.824 LEC1 MapReduce 笔记