TPC-H#

The TPC-H is a decision support benchmark. It consists of a suite of business oriented ad-hoc queries and concurrent data modifications. The queries and the data populating the database have been chosen to have broad industry-wide relevance.

Tip

This article assumes that you have mastered the basic knowledge and operation of TPC-H. For the knowledge about TPC-H not mentioned in this article, you can obtain it from its Official Documentation.

This connector can be used to test the capabilities and query syntax of Spark without configuring access to an external data source. When you query a TPC-H table, the connector generates the data on the fly using a deterministic algorithm.

Goto Try Kyuubi to explore TPC-H data instantly!

TPC-H Integration#

To enable the integration of kyuubi spark sql engine and TPC-H through Apache Spark Datasource V2 and Catalog APIs, you need to:

Dependencies#

The classpath of kyuubi spark sql engine with TPC-H supported consists of

  1. kyuubi-spark-sql-engine-1.7.0-SNAPSHOT_2.12.jar, the engine jar deployed with Kyuubi distributions

  2. a copy of spark distribution

  3. kyuubi-spark-connector-tpch-1.7.0-SNAPSHOT_2.12.jar, which can be found in the Maven Central

In order to make the TPC-H connector package visible for the runtime classpath of engines, we can use one of these methods:

  1. Put the TPC-H connector package into $SPARK_HOME/jars directly

  2. Set spark.jars=kyuubi-spark-connector-tpch-1.7.0-SNAPSHOT_2.12.jar

Configurations#

To add TPC-H tables as a catalog, we can set the following configurations in $SPARK_HOME/conf/spark-defaults.conf:

# (required) Register a catalog named `tpch` for the spark engine.
spark.sql.catalog.tpch=org.apache.kyuubi.spark.connector.tpch.TPCHCatalog

# (optional) Excluded database list from the catalog, all available databases are:
#            sf0, tiny, sf1, sf10, sf30, sf100, sf300, sf1000, sf3000, sf10000, sf30000, sf100000.
spark.sql.catalog.tpch.excludeDatabases=sf10000,sf30000

# (optional) When true, use CHAR/VARCHAR, otherwise use STRING. It affects output of the table schema,
#            e.g. `SHOW CREATE TABLE <table>`, `DESC <table>`.
spark.sql.catalog.tpch.useAnsiStringType=false

# (optional) Maximum bytes per task, consider reducing it if you want higher parallelism.
spark.sql.catalog.tpch.read.maxPartitionBytes=128m

TPC-H Operations#

Listing databases under tpch catalog.

SHOW DATABASES IN tpch;

Listing tables under tpch.sf1 database.

SHOW TABLES IN tpch.sf1;

Switch current database to tpch.sf1 and run a query against it.

USE tpch.sf1;
SELECT * FROM orders;