sunjincheng121 commented on a change in pull request #817: [IOTDB-497] Apache Flink Connector Support URL: https://github.com/apache/incubator-iotdb/pull/817#discussion_r382860039
########## File path: flink-iotdb-connector/README.md ########## @@ -0,0 +1,65 @@ +# IoTDB-Flink-Connector + +IoTDB integration for [Apache Flink](https://flink.apache.org/). This module includes the iotdb sink that allows a flink job to write events into timeseries. + +## IoTDBSink +To use the `IoTDBSink`, you need construct an instance of it by specifying `IoTDBOptions` and `IoTSerializationSchema` instances. +The `IoTDBSink` send only one event after another by default, but you can change to batch by invoking `withBatchFlushOnCheckpoint(true)`. + +## Examples +The following is an example which receiving events from sensor source and then sending events to iotdb. + + ```java +StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment(); + env.enableCheckpointing(3000); + + IoTDBOptions options = new IoTDBOptions(); + options.setHost("127.0.0.1"); + options.setPort(6667); + options.setUser("root"); + options.setPassword("root"); + options.setStorageGroup("root.sg"); + options.setTimeseries(Lists.newArrayList("root.sg.d1.s1")); + + IoTSerializationSchema serializationSchema = new DefaultIoTSerializationSchema(); + IoTDBSink ioTDBSink = new IoTDBSink(options, serializationSchema) + // enable batching + .withBatchFlushOnCheckpoint(true) + .withBatchSize(10) Review comment: What if it is set `batchsize` but not turned on `BatchFlushOnCheckpoint`? I mean that user always want enable batching if user set `batchsize`, if so, do we really need `withBatchFlushOnCheckpoint`? ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
