In the previous blog, we learned Flink's File Sink. Interested students can refer to the following:
Flink tutorial (01) - Flink knowledge mapFlink tutorial (02) - getting started with FlinkFlink tutorial (03) - Flink environment constructionFlink tutorial (04) - getting started with FlinkFlink tutorial (05) - simple analys ...
Added by !Mikey on Wed, 09 Mar 2022 03:29:25 +0200
This article was first published in Floating finch language:https://www.yuque.com/17singeditiondateremarks1.02022.3.8Article launchThis paper analyzes the code based on Flink 1.14.0. PrefaceA while ago, I encountered a strange phenomenon in production: the full-scale operation cannot be carried out normally, and the log is full of Java util. co ...
Added by nyfael on Wed, 09 Mar 2022 02:03:12 +0200
1. Operator state classification
The scope of action of operator state is limited to operator parallel subtasks. This means that all data processed by the same parallel subtask can access the same state, which is shared for the same subtask. Operator state cannot be accessed by another parallel subtask of the same or different operators.
Added by Qlubbie on Tue, 08 Mar 2022 04:59:25 +0200
In the previous blog, we have a certain understanding of the use of Source in Flink's program model. Interested students can refer to the following:
Flink tutorial (01) - Flink knowledge mapFlink tutorial (02) - getting started with FlinkFlink tutorial (03) - Flink environment constructionFlink tutorial (04) - getting star ...
Added by journeyman73 on Mon, 07 Mar 2022 10:24:51 +0200
By default, when the watermark passes through the end of window and the previous data arrives, these data will be deleted.
In order to avoid some late data being deleted, the concept of allowed lateness is generated.
In short, allowed latency is for event time. After the watermark exceeds the end of window, it is also allowed to wait for a pe ...
Added by vargadanis on Thu, 24 Feb 2022 13:31:24 +0200
Windows is the core of handling infinite streams. Windows divides streams into "buckets" of limited size, on which we can apply calculations. This document focuses on how to perform windowing in Flink SQL and how programmers can get the most from the functions they provide.
Apache Flink provides several window table ...
Added by mattsutton on Wed, 23 Feb 2022 18:08:35 +0200
In the previous article, we introduced the installation, deployment and basic concepts of Flink. Today, let's learn about DataStream API, one of the core of Flink.
01 distributed stream processing foundation
In the figure above, we divide the whole code into three parts, namely, the basic model of distributed stream processing:
Added by c-o-d-e on Sun, 20 Feb 2022 10:32:54 +0200
Use Docker to configure flink1 12 development test environment
Docker is now an indispensable tool in the development community. We can quickly test and package our applications with docker. At the same time, using containers, we can easily realize the deployment and configuration of CI and CD in software development. At ...
Added by steve490 on Fri, 18 Feb 2022 17:44:04 +0200