发布于 2017-12-14 00:16:56 | 200 次阅读 | 评论: 0 | 来源: 网友投递
Apache Flink 通用数据处理平台
Apache Flink 声明式的数据分析开源系统,结合了分布式 MapReduce 类平台的高效,灵活的编程和扩展性。同时在并行数据库发现查询优化方案。
Apache Flink 1.4.0 发布了。ApacheFlink 是一个开源的流处理框架,应用于分布式,高性能,始终可用的,准确的数据流应用程序。
此版本更新内容:
[FLINK-3551] - Sync Scala and Java Streaming Examples
[FLINK-3582] - Add Iterator over State for All Keys in Partitioned State
[FLINK-5270] - Refactor the batch Scala-expression Table API tests
[FLINK-5272] - Refactor the stream Scala-expression Table API tests
[FLINK-5509] - Replace QueryableStateClient keyHashCode argument
[FLINK-5604] - Replace QueryableStateClient constructor
[FLINK-3182] - TaskManagersHandler can block web frontend
[FLINK-4047] - Fix documentation about determinism of KeySelectors
[FLINK-4048] - Remove Hadoop dependencies from ExecutionEnvironment
[FLINK-4228] - YARN artifact upload does not work with S3AFileSystem
[FLINK-4500] - Cassandra sink can lose messages
[FLINK-4660] - HadoopFileSystem (with S3A) may leak connections, which cause job to stuck in a restarting loop
[FLINK-5070] - Unable to use Scala's BeanProperty with classes
该版本可从以下网址下载: https://flink.apache.org/downloads.html