-
Concurrency Control
Deployment models with supported concurrency controls Model A: Single writer with inline table services Single Writer Guarantees Model B: Single writer with async table services ... -
本地参数
本地参数 作用域 在任务定义页面配置的参数，默认作用域仅限该任务，如果配置了参数传递 则可将该参数作用到下游任务中。 使用方式 本地参数配置方式如下：在任务定义页面，点击“自定义参数”右边的加号，填写对应的变量名称和对应的值，保存即可。 如果要在单个任务中使用参数，请参阅通过自定义参数使用 如果要在任务中使用配置参数并在下游任务中使用它们： 如果... -
Kafka
Support Those Engines Key Features Description Supported DataSource Info Sink Options Parameter Interpretation Topic Formats Semantics Partition Key Fields Assign Partition... -
S3File
Support Those Engines Key Features Description Supported DataSource Info Database Dependency Data Type Mapping Orc File Type Parquet File Type Sink Options path [string] h... -
Flink Quick Start
Setup Flink Support Matrix Download Flink and Start Flink cluster Start Flink SQL client Create Table Insert Data Query Data Update Data Delete Data Row-level Delete Batch... -
Utility
noConflict identity constant noop times random mixin iteratee uniqueId escape unescape result now template noConflict _.noConflict() source Give control of the... -
Accumulo Clients
Creating Client Code Creating an Accumulo Client Authentication Writing Data BatchWriter ConditionalWriter Durability Reading Data Scanner Isolated Scanner BatchScanner ... -
Clustering
Background How is compaction different from clustering? Clustering Architecture Overall, there are 2 steps to clustering Schedule clustering Execute clustering Clustering Use... -
6. Distributed objects
6.1. Object holder 6.1.1. Object holder listeners 6.2. Binary stream holder 6.2.1. Binary stream holder listeners 6.3. Geospatial holder 6.4. BitSet 6.4.1. BitSet data partiti... -
Design & Concepts
How does Hudi ensure atomicity? Does Hudi extend the Hive table layout? What concurrency control approaches does Hudi adopt? Hudi’s commits are based on transaction start time i...