当前位置:网站首页>Flinksql window triggered in advance
Flinksql window triggered in advance
2022-07-27 01:14:00 【Me fan】
scene : The window of the day , Want to trigger every minute or other time point
insert into sliding_window_local_pay_day
select userid,
TUMBLE_START (ptime, INTERVAL '1' DAY) as window_start,
TUMBLE_END (ptime, INTERVAL '1' DAY) as window_end,
COUNT (1) as pay_num
from
flink_kafka_join_pay
group by TUMBLE (ptime, INTERVAL '1' DAY), userid;1. Two parameters : Every minute
table.exec.emit.early-fire.enabled: 'true'
table.exec.emit.early-fire.delay: 60s 2. Triggering in the source code means generating Trigger
Generate Trigger The logic of is window aggregation key The first data time point +N individual Interval, It's different key The trigger time is different
/**
* Creates a trigger that fires by a certain interval after reception of the first element.
*
* @param time the certain interval
*/
public static <W extends Window> AfterFirstElementPeriodic<W> every(Duration time) {
return new AfterFirstElementPeriodic<>(time.toMillis());
}3. Trigger effect , Different userid, The trigger time is different

边栏推荐
- 7. F1方程式冠军
- 并发编程之生产者消费者模式
- Write the changed data in MySQL to Kafka through flinkcdc (datastream mode)
- 李宏毅机器学习(2017版)_P13:深度学习
- Rational selection of (Spark Tuning ~) operator
- Spark ---- shuffle and partition of RDD
- MySQL split table DDL operation (stored procedure)
- adb shell截屏录屏命令
- 李宏毅机器学习(2017版)_P5:误差
- More than live streaming: what other eye-catching functions does Tencent cloud live mlvb plug-in have besides streaming / streaming
猜你喜欢

Android——LitePal数据库框架的基本用法

李宏毅机器学习(2017版)_P6-8:梯度下降

Scala-模式匹配

Real time calculation demo based on Flink: user behavior analysis (IV: how many different users have visited the website (UV) in a period of time)

MTCNN

Deep understanding of golang - closures

什么是数字经济,它是如何改变商业模式的?

短视频App开发有哪些必备的功能?

Solve the problem that rsyslog service occupies too much memory

Li Hongyi machine learning (2017 Edition)_ P1-2: introduction to machine learning
随机推荐
Android——LitePal数据库框架的基本用法
Real time calculation demo based on Flink: user behavior analysis (IV: how many different users have visited the website (UV) in a period of time)
FaceNet
Small programs related to a large number of digital collections off the shelves of wechat: is NFT products the future or a trap?
使用tika 判断文件类型
Li Hongyi machine learning (2017 Edition)_ P14: back propagation
Redisson working principle - source code analysis
Website log collection and analysis process
DataNode Decommision
Status management in Flink
In depth learning report (3)
Iptables 详解与实战案例
视频类小程序变现的最短路径:从带货到品牌营销
基于485总线的评分系统
非递归前中后序遍历二叉树
Kubernetes 是什么 ?
Spark ---- shuffle and partition of RDD
1. 众数
物联网将彻底改变制造业,你准备好了吗?
智密-腾讯云直播 MLVB 插件优化教程:六步提升拉流速度+降低直播延迟