Apache Spark - Apache HBase Connector
The Apache Spark - Apache HBase Connector is a library to support Spark accessing HBase table as external data source or sink. With it, user can operate HBase with Spark-SQL on DataFrame and DataSet level.
With the DataFrame and DataSet support, the library leverages all the optimization techniques in catalyst, and achieves data locality, partition pruning, predicate pushdown, Scanning and BulkGet, etc.
Catalog
For each table, a catalog has to be provided, which includes the row key, and the columns with data type with predefined column families, and defines the mapping between hbase column and table schema. The catalog is user defined json format.
Datatype conversion
Java primitive types is supported. In the future, other data types will be supported, which relies on user specified serdes. There are three internal serdes supported in SHC: Avro, Phoenix, PrimitiveType. User can specify which serde they want to use by defining 'tableCoder' in their catalog. For this, please refer to examples and unit tests. Take Avro as an example. User defined serdes will be responsible to convert byte array to Avro object, and connector will be responsible to convert Avro object to catalyst supported data types. When user define a new serde, they need to make it 'implement' the trait 'SHCDataType'.
Note that if user want DataFrame to only handle byte array, the binary type can be specified. Then user can get the catalyst row with each column as a byte array. User can further deserialize it with customized deserializer, or operate on the RDD of the DataFrame directly.
Data locality
When the spark worker nodes are co-located with hbase region servers, data locality is achieved by identifying the region server location, and co-locate the executor with the region server. Each executor will only perform Scan/BulkGet on the part of the data that co-locates on the same host.
Predicate pushdown
The library uses existing standard HBase filter provided by HBase and does not operate on the coprocessor.
Partition Pruning
By extracting the row key from the predicates, we split the scan/BulkGet into multiple non-overlapping regions, only the region servers that have the requested data will perform scan/BulkGet. Currently, the partition pruning is performed on the first dimension of the row keys. Note that the WHERE conditions need to be defined carefully. Otherwise, the result scanning may includes a region larger than user expectd. For example, following condition will result in a full scan (rowkey1 is the first dimension of the rowkey, and column is a regular hbase column). WHERE rowkey1 > "abc" OR column = "xyz"
Scanning and BulkGet
Both are exposed to users by specifying WHERE CLAUSE, e.g., where column > x and column < y for scan and where column = x for get. All the operations are performed in the executors, and driver only constructs these operations. Internally we will convert them to scan or get or combination of both, which return Iterator[Row] to catalyst engine.
Creatable DataSource
The libary support both read/write from/to HBase.
Compile
mvn package -DskipTests
Running Tests and Examples
Run test
mvn clean package test
Run indiviudal test
mvn -DwildcardSuites=org.apache.spark.sql.DefaultSourceSuite test
Run SHC examples
./bin/spark-submit --verbose --class org.apache.spark.sql.execution.datasources.hbase.examples.HBaseSource --master yarn-cluster --packages com.hortonworks:shc-core:1.1.1-2.1-s_2.11 --repositories http://repo.hortonworks.com/content/groups/public/ --files /usr/hdp/current/hbase-client/conf/hbase-site.xml shc-examples-1.1.1-2.1-s_2.11-SNAPSHOT.jar
The following illustrates how to run your application in real hbase cluster. You need to provide the hbase-site.xml. It may subject to change based on your specific cluster configuration.
./bin/spark-submit --class your.application.class --master yarn-client --packages com.hortonworks:shc-core:1.1.1-2.1-s_2.11 --repositories http://repo.hortonworks.com/content/groups/public/ --files /etc/hbase/conf/hbase-site.xml /To/your/application/jar
Running Spark applications with this connector, HBase jars of version 1.1.2 will be pulled by default. If Phoenix is enabled on HBase cluster, you need to use "--jars" to pass "phoenix-server.jar". For example:
./bin/spark-submit --class your.application.class --master yarn-client --packages com.hortonworks:shc-core:1.1.1-2.1-s_2.11 --repositories http://repo.hortonworks.com/content/groups/public/ --jars /usr/hdp/current/phoenix-client/phoenix-server.jar --files /etc/hbase/conf/hbase-site.xml /To/your/application/jar
Application Usage
The following illustrates the basic procedure on how to use the connector. For more details and advanced use case, such as Avro and composite key support, please refer to the examples in the repository.
Defined the HBase catalog
def catalog = s"""{
|"table":{"namespace":"default", "name":"table1"},
|"rowkey":"key",
|"columns":{
|"col0":{"cf":"rowkey", "col":"key", "type":"string"},
|"col1":{"cf":"cf1", "col":"col1", "type":"boolean"},
|"col2":{"cf":"cf2", "col":"col2", "type":"double"},
|"col3":{"cf":"cf3", "col":"col3", "type":"float"},
|"col4":{"cf":"cf4", "col":"col4", "type":"int"},
|"col5":{"cf":"cf5", "col":"col5", "type":"bigint"},
|"col6":{"cf":"cf6", "col":"col6", "type":"smallint"},
|"col7":{"cf":"cf7", "col":"col7", "type":"string"},
|"col8":{"cf":"cf8", "col":"col8", "type":"tinyint"}
|}
|}""".stripMargin
The above defines a schema for a HBase table with name as table1, row key as key and a number of columns (col1-col8). Note that the rowkey also has to be defined in details as a column (col0), which has a specific cf (rowkey).
Write to HBase table to populate data
sc.parallelize(data).toDF.write.options(
Map(HBaseTableCatalog.tableCatalog -> catalog, HBaseTableCatalog.newTable -> "5"))
.format("org.apache.spark.sql.execution.datasources.hbase")
.save()
Given a DataFrame with specified schema, above will create an HBase table with 5 regions and save the DataFrame inside. Note that if HBaseTableCatalog.newTable is not specified, the table has to be pre-created.
Perform DataFrame operation on top of HBase table
def withCatalog(cat: String): DataFrame = {
sqlContext
.read
.options(Map(HBaseTableCatalog.tableCatalog->cat))
.format("org.apache.spark.sql.execution.datasources.hbase")
.load()
}
Complicated query
val df = withCatalog(catalog)
val s = df.filter((($"col0" <= "row050" && $"col0" > "row040") ||
$"col0" === "row005" ||
$"col0" === "row020" ||
$"col0" === "r20" ||
$"col0" <= "row005") &&
($"col4" === 1 ||
$"col4" === 42))
.select("col0", "col1", "col4")
s.show
SQL support
// Load the dataframe
val df = withCatalog(catalog)
//SQL example
df.createOrReplaceTempView("table")
sqlContext.sql("select count(col1) from table").show
Configuring Spark-package
Users can use the Spark-on-HBase connector as a standard Spark package. To include the package in your Spark application use:
Note: com.hortonworks:shc-core:1.1.1-2.1-s_2.11 has not been uploaded to spark-packages.org, but will be there soon.
spark-shell, pyspark, or spark-submit
$SPARK_HOME/bin/spark-shell --packages com.hortonworks:shc-core:1.1.1-2.1-s_2.11
Users can include the package as the dependency in your SBT file as well. The format is the spark-package-name:version in build.sbt file.
libraryDependencies += “com.hortonworks/shc-core:1.1.1-2.1-s_2.11”
Running in secure cluster
For running in a Kerberos enabled cluster, the user has to include HBase related jars into the classpath as the HBase token retrieval and renewal is done by Spark, and is independent of the connector. In other words, the user needs to initiate the environment in the normal way, either through kinit or by providing principal/keytab. The following examples show how to run in a secure cluster with both yarn-client and yarn-cluster mode. Note that if your Spark does not contain SPARK-20059, which is in Apache Spark 2.1.1+, and SPARK-21377, which is in Apache Spark 2.3.0+, you need to set SPARK_CLASSPATH for both modes (refer here).
Suppose hrt_qa is a headless account, user can use following command for kinit:
kinit -k -t /tmp/hrt_qa.headless.keytab hrt_qa
/usr/hdp/current/spark-client/bin/spark-submit --class your.application.class --master yarn-client --files /etc/hbase/conf/hbase-site.xml --packages com.hortonworks:shc-core:1.1.1-2.1-s_2.11 --repositories http://repo.hortonworks.com/content/groups/public/ /To/your/application/jar
/usr/hdp/current/spark-client/bin/spark-submit --class your.application.class --master yarn-cluster --files /etc/hbase/conf/hbase-site.xml --packages com.hortonworks:shc-core:1.1.1-2.1-s_2.11 --repositories http://repo.hortonworks.com/content/groups/public/ /To/your/application/jar
If the solution above does not work and you encounter errors like :
org.apache.zookeeper.ZooKeeper: Initiating client connection, connectString=localhost:2181
or
ERROR ipc.AbstractRpcClient: SASL authentication failed. The most likely cause is missing or invalid credentials. Consider 'kinit'.
javax.security.sasl.SaslException: GSS initiate failed [Caused by GSSException: No valid credentials provided (Mechanism level: Failed to find any Kerberos tgt)]
Include the hbase-site.xml under SPARK_CONF_DIR (/etc/spark/conf) on the host where the spark job is submitted from, by creating a symbolic link towards your main hbase-site.xml (in order to be synchronous with your platform updates).
Using SHCCredentialsManager
Spark only supports use cases which access a single secure HBase cluster. If your applications need to access multiple secure HBase clusters, users need to use SHCCredentialsManager instead. SHCCredentialsManager supports a single secure HBase cluster as well as multiple secure HBase clusters. It is disabled by default, but users can set spark.hbase.connector.security.credentials.enabled to true to enable it. Also, users need to config principal and keytab as below before running their applications.
spark.hbase.connector.security.credentials.enabled true
spark.hbase.connector.security.credentials ambari-qa-c1@EXAMPLE.COM
spark.hbase.connector.security.keytab /etc/security/keytabs/smokeuser.headless.keytab
or
spark.hbase.connector.security.credentials.enabled true
spark.yarn.principal ambari-qa-c1@EXAMPLE.COM
spark.yarn.keytab /etc/security/keytabs/smokeuser.headless.keytab
Others
Example. Support of Avro schemas:
The connector fully supports all the avro schemas. Users can use either a complete record schema or partial field schema as data type in their catalog (refer here for more detailed information).
val schema_array = s"""{"type": "array", "items": ["string","null"]}""".stripMargin
val schema_record =
s"""{"namespace": "example.avro",
| "type": "record", "name": "User",
| "fields": [ {"name": "name", "type": "string"},
| {"name": "favorite_number", "type": ["int", "null"]},
| {"name": "favorite_color", "type": ["string", "null"]} ] }""".stripMargin
val catalog = s"""{
|"table":{"namespace":"default", "name":"htable"},
|"rowkey":"key1",
|"columns":{
|"col1":{"cf":"rowkey", "col":"key1", "type":"double"},
|"col2":{"cf":"cf1", "col":"col1", "avro":"schema_array"},
|"col3":{"cf":"cf1", "col":"col2", "avro":"schema_record"},
|"col4":{"cf":"cf1", "col":"col3", "type":"double"},
|"col5":{"cf":"cf1", "col":"col4", "type":"string"}
|}
|}""".stripMargin
val df = sqlContext.read.options(Map("schema_array"->schema_array,"schema_record"->schema_record, HBaseTableCatalog.tableCatalog->catalog)).format("org.apache.spark.sql.execution.datasources.hbase").load()
df.write.options(Map("schema_array"->schema_array,"schema_record"->schema_record, HBaseTableCatalog.tableCatalog->catalog)).format("org.apache.spark.sql.execution.datasources.hbase").save()
TODO:
val complex = s"""MAP<int, struct<varchar:string>>"""
val schema =
s"""{"namespace": "example.avro",
| "type": "record", "name": "User",
| "fields": [ {"name": "name", "type": "string"},
| {"name": "favorite_number", "type": ["int", "null"]},
| {"name": "favorite_color", "type": ["string", "null"]} ] }""".stripMargin
val catalog = s"""{
|"table":{"namespace":"default", "name":"htable"},
|"rowkey":"key1:key2",
|"columns":{
|"col1":{"cf":"rowkey", "col":"key1", "type":"binary"},
|"col2":{"cf":"rowkey", "col":"key2", "type":"double"},
|"col3":{"cf":"cf1", "col":"col1", "avro":"schema1"},
|"col4":{"cf":"cf1", "col":"col2", "type":"string"},
|"col5":{"cf":"cf1", "col":"col3", "type":"double", "sedes":"org.apache.spark.sql.execution.datasources.hbase.DoubleSedes"},
|"col6":{"cf":"cf1", "col":"col4", "type":"$complex"}
|}
|}""".stripMargin
val df = sqlContext.read.options(Map("schema1"->schema, HBaseTableCatalog.tableCatalog->catalog)).format("org.apache.spark.sql.execution.datasources.hbase").load()
df.write.options(Map("schema1"->schema, HBaseTableCatalog.tableCatalog->catalog)).format("org.apache.spark.sql.execution.datasources.hbase").save()
Above illustrates our next step, which includes composite key support, complex data types, support of customerized serde and avro. Note that although all the major pieces are included in the current code base, but it may not be functioning now.
Trademarks
Apache®, Apache Spark, Apache HBase, Spark, and HBase are either registered trademarks or trademarks of the Apache Software Foundation in the United States and/or other countries.