rdblue commented on a change in pull request #1100: URL: https://github.com/apache/iceberg/pull/1100#discussion_r445246402
########## File path: docker/README.md ########## @@ -0,0 +1,118 @@ +<!-- + - Licensed to the Apache Software Foundation (ASF) under one + - or more contributor license agreements. See the NOTICE file + - distributed with this work for additional information + - regarding copyright ownership. The ASF licenses this file + - to you under the Apache License, Version 2.0 (the + - "License"); you may not use this file except in compliance + - with the License. You may obtain a copy of the License at + - + - http://www.apache.org/licenses/LICENSE-2.0 + - + - Unless required by applicable law or agreed to in writing, + - software distributed under the License is distributed on an + - "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + - KIND, either express or implied. See the License for the + - specific language governing permissions and limitations + - under the License. + --> + +# Version + +| Framework | Version | +| :--------- | :------- | +| Hadoop | 2.7.4 | +| Spark | 2.4.5 | +| Iceberg | 0.8.0 | + +# Setting up Docker Demo + +```$xslt +# start the docker demo +cd docker +./start_demo.sh + +# You can see the output below if the docker demo starts successfully +Creating network "compose_default" with the default driver +Creating namenode ... done +Creating datanode ... done +Creating spark ... done +``` + +# Demo + +At this point, you can enter the container and try something with Spark and Iceberg. + +```$xslt +docker exec -it spark /bin/bash + +# After getting into the container, we start with spark-shell +spark-shell --master local[2] +``` + +## Test Data + +You can check the data located on /opt/data/logs.json, which shows a few records of logging data. + +```$xslt +{"level": "INFO", "event_time": 1591430621, "message": "Containers are ready.", "call_stack": []} +{"level": "INFO", "event_time": 1591430621, "message": "Start working with Iceberg!", "call_stack": []} +{"level": "WARN", "event_time": 1591430621, "message": "This is a warn meesage", "call_stack": ["functionA", "functionB"]} +{"level": "ERROR", "event_time": 1591430621, "message": "NullPointerException", "call_stack": ["String.substring(int, int)"]} +{"level": "ERROR", "event_time": 1591430621, "message": "IllegalArgumentException", "call_stack": ["unknow stack"]} +{"level": "INFO", "event_time": 1591430621, "message": "The cluster is shutting donw.", "call_stack": []} +``` + +## Create an Iceberg Table + +Let's start with creating an Iceberg table on HDFS. + +```$xslt +import org.apache.iceberg.Schema +import org.apache.iceberg.types.Types._ + +val schema = new Schema( + NestedField.required(1, "level", StringType.get()), + NestedField.required(2, "event_time", TimestampType.withZone()), + NestedField.required(3, "message", StringType.get()), + NestedField.optional(4, "call_stack", ListType.ofRequired(5, StringType.get())) +) + +import org.apache.iceberg.PartitionSpec +val spec = PartitionSpec.builderFor(schema).hour("event_time").build() + +import org.apache.iceberg.hadoop.HadoopTables +val tables = new HadoopTables(spark.sessionState.newHadoopConf()) +val table = tables.create(schema, spec, "hdfs:/tables/logging/logs") +``` + +## Load Data + +The data can loaded as a dataframe with Spark.(We use sparkSchema here because the schema of the dataframe should be +compatitable with the Iceberg schema we used above.) + +```$xslt +import org.apache.spark.sql.types._ +val sparkSchema = StructType(Seq( + StructField("level", StringType, nullable = false), + StructField("event_time", TimestampType, nullable = false), + StructField("message", StringType, nullable = false), + StructField("call_stack", ArrayType(StringType, containsNull = false), nullable = false) +)) +val logsDF = spark.read.schema(sparkSchema).json("file:///opt/data/logs.json") +val df = spark.createDataFrame(logsDF.rdd, sparkSchema) +``` + +## Write Data + +```$xslt +df.write.format("iceberg").mode("append").save("hdfs:/tables/logging/logs") Review comment: I'd recommend adding a sort to the DataFrame. Iceberg doesn't allow un-grouped data to be written out -- that is, if a task writes a file for a given partition (hour), then all of the rows for that partition need to be grouped together. Sorting handles that and solves a lot of other Spark problems. ```scala df.orderBy("event_time").write... ``` ########## File path: docker/README.md ########## @@ -0,0 +1,118 @@ +<!-- + - Licensed to the Apache Software Foundation (ASF) under one + - or more contributor license agreements. See the NOTICE file + - distributed with this work for additional information + - regarding copyright ownership. The ASF licenses this file + - to you under the Apache License, Version 2.0 (the + - "License"); you may not use this file except in compliance + - with the License. You may obtain a copy of the License at + - + - http://www.apache.org/licenses/LICENSE-2.0 + - + - Unless required by applicable law or agreed to in writing, + - software distributed under the License is distributed on an + - "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + - KIND, either express or implied. See the License for the + - specific language governing permissions and limitations + - under the License. + --> + +# Version + +| Framework | Version | +| :--------- | :------- | +| Hadoop | 2.7.4 | +| Spark | 2.4.5 | +| Iceberg | 0.8.0 | + +# Setting up Docker Demo + +```$xslt +# start the docker demo +cd docker +./start_demo.sh + +# You can see the output below if the docker demo starts successfully +Creating network "compose_default" with the default driver +Creating namenode ... done +Creating datanode ... done +Creating spark ... done +``` + +# Demo + +At this point, you can enter the container and try something with Spark and Iceberg. + +```$xslt +docker exec -it spark /bin/bash + +# After getting into the container, we start with spark-shell +spark-shell --master local[2] +``` + +## Test Data + +You can check the data located on /opt/data/logs.json, which shows a few records of logging data. + +```$xslt +{"level": "INFO", "event_time": 1591430621, "message": "Containers are ready.", "call_stack": []} +{"level": "INFO", "event_time": 1591430621, "message": "Start working with Iceberg!", "call_stack": []} +{"level": "WARN", "event_time": 1591430621, "message": "This is a warn meesage", "call_stack": ["functionA", "functionB"]} +{"level": "ERROR", "event_time": 1591430621, "message": "NullPointerException", "call_stack": ["String.substring(int, int)"]} +{"level": "ERROR", "event_time": 1591430621, "message": "IllegalArgumentException", "call_stack": ["unknow stack"]} +{"level": "INFO", "event_time": 1591430621, "message": "The cluster is shutting donw.", "call_stack": []} +``` + +## Create an Iceberg Table + +Let's start with creating an Iceberg table on HDFS. + +```$xslt +import org.apache.iceberg.Schema +import org.apache.iceberg.types.Types._ + +val schema = new Schema( + NestedField.required(1, "level", StringType.get()), + NestedField.required(2, "event_time", TimestampType.withZone()), + NestedField.required(3, "message", StringType.get()), + NestedField.optional(4, "call_stack", ListType.ofRequired(5, StringType.get())) +) + +import org.apache.iceberg.PartitionSpec +val spec = PartitionSpec.builderFor(schema).hour("event_time").build() + +import org.apache.iceberg.hadoop.HadoopTables +val tables = new HadoopTables(spark.sessionState.newHadoopConf()) +val table = tables.create(schema, spec, "hdfs:/tables/logging/logs") +``` + +## Load Data + +The data can loaded as a dataframe with Spark.(We use sparkSchema here because the schema of the dataframe should be +compatitable with the Iceberg schema we used above.) + +```$xslt Review comment: Why xslt? ---------------------------------------------------------------- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
