Index TypesÂ¶

Currently, Tajo supports only one type of index, TWO_LEVEL_BIN_TREE, shortly BST. The BST index is a kind of binary search tree which is extended to be permanently stored on disk. It consists of two levels of nodes; a leaf node indexes the keys with the offsets to data stored on HDFS, and a root node indexes the keys with the offsets to the leaf nodes.

When an index scan is started, the query engine first reads the root node and finds the search key. If it successfully finds a leaf node corresponding to the search key, it subsequently finds the search key in that leaf node. Finally, it directly reads a tuple corresponding to the search key from HDFS.

Index (Experimental Feature)Â¶

An index is a data structure that is used for efficient query processing. Using an index, the Tajo query engine can directly retrieve search values.

The following sections describe the supported index types, the query execution with an index, and the future works.

IntroductionÂ¶

The main goal of Apache Tajo project is to build an advanced open source +data warehouse system in Hadoop for processing web-scale data sets. +Basically, Tajo provides SQL standard as a query language. +Tajo is designed for both interactive and batch queries on data sets +stored on HDFS and other data sources. Without hurting query response +times, Tajo provides fault-tolerance and dynamic load balancing which +are necessary for long-running queries. Tajo employs a cost-based and +progressive query optimization techniques for reoptimizing running +queries in order to avoid the worst query plans.

+ +

Tajo JDBC DriverÂ¶

Apache Tajoâ¢ provides JDBC driver +which enables Java applciations to easily access Apache Tajo in a RDBMS-like manner. +In this section, we explain how to get JDBC driver and an example code.

How to get JDBC driverÂ¶

Direct DownloadÂ¶

You can directly download a JDBC driver jar file (tajo-jdbc-x.y.z.jar) from Downloads.

From Binary DistributionÂ¶

Tajo binary distribution provides JDBC jar file located in ${TAJO_HOME}/share/jdbc-dist/tajo-jdbc-x.y.z.jar.

From Building Source CodeÂ¶

You can build Tajo from the source code and then get JAR files as follows:

$ tar xzvf tajo-x.y.z-src.tar.gz
+$ mvn clean package -DskipTests -Pdist -Dtar
+$ ls -l tajo-dist/target/tajo-x.y.z/share/jdbc-dist/tajo-jdbc-x.y.z.jar
+

+

Setting the CLASSPATHÂ¶

In order to use the JDBC driver, you should add tajo-jdbc-x.y.z.jar in your CLASSPATH.

CLASSPATH=path/to/tajo-jdbc-x.y.z.jar:$CLASSPATH
+

+

Connecting to the Tajo cluster instanceÂ¶

A Tajo cluster is represented by a URL. Tajo JDBC driver can take the following URL forms:

+
+
jdbc:tajo://host/
+
jdbc:tajo://host/database
+
jdbc:tajo://host:port/
+
jdbc:tajo://host:port/database
+
+

Each part of URL has the following meanings:

+
+
host - The hostname of the TajoMaster. You can put hostname or ip address here.
+
port - The port number that server is listening. Default port number is 26002.
+
database - The database name. The default database name is default.
+
+
To connect, you need to get Connection instance from Java JDBC Driver Manager as follows:
+

Connection db = DriverManager.getConnection(url);
+

+

Connection ParametersÂ¶

Connection parameters lets the JDBC Copnnection to enable or disable additional features. You should use java.util.Properties to pass your connection parameters into Connection. The following example means that the transmission of ResultSet uses compression and its connection timeout is 15 seconds.

String url = "jdbc:tajo://localhost/test";
+Properties props = new Properties();
+props.setProperty("useCompression","true");  // use compression for ResultSet
+props.setProperty("connectTimeout","15000"); // 15 seconds
+Connection conn = DriverManager.getConnection(url, props);
+

+

The connection parameters that Tajo currently supports are as follows:

+
+
useCompression = bool - Enable compressed transfer for ResultSet.
+
defaultRowFetchSize = int - Determine the number of rows fetched in ResultSet by one fetch with trip to the Server.
+
connectTimeout = int (seconds) - The timeout value used for socket connect operations. If connecting to the server takes longer than this value, the connection is broken. The timeout is specified in seconds and a value of zero means that it is disabled.
+
socketTimeout = int (seconds) - The timeout value used for socket read operations. If reading from the server takes longer than this value, the connection is closed. This can be used as both a brute force global query timeout and a method of detecting network problems. The timeout is specified in seconds and a value of zero means that it is disabled.
+
retry = int - Number of retry operation. Tajo JDBC driver is resilient against some network or connection problems. It determines how many times the connection will retry.
+
+

An Example JDBC ClientÂ¶

The JDBC driver class name is org.apache.tajo.jdbc.TajoDriver. +You can get the driver Class.forName("org.apache.tajo.jdbc.TajoDriver"). +The connection url should be jdbc:tajo://<TajoMaster hostname>:<TajoMaster client rpc port>/<database name>. +The default TajoMaster client rpc port is 26002. +If you want to change the listening port, please refer Cluster Setup.

Note

Currently, Tajo does not support the concept of database and namespace. +All tables are contained in default database. So, you don’t need to specify any database name.

The following shows an example of JDBC Client.

import java.sql.Connection;
+import java.sql.ResultSet;
+import java.sql.Statement;
+import java.sql.DriverManager;
+
+public class TajoJDBCClient {
+
+  ....
+
+  public static void main(String[] args) throws Exception {
+
+    try {
+      Class.forName("org.apache.tajo.jdbc.TajoDriver");
+    } catch (ClassNotFoundException e) {
+      // fill your handling code
+    }
+
+    Connection conn = DriverManager.getConnection("jdbc:tajo://127.0.0.1:26002/default");
+
+    Statement stmt = null;
+    ResultSet rs = null;
+    try {
+      stmt = conn.createStatement();
+      rs = stmt.executeQuery("select * from table1");
+      while (rs.next()) {
+        System.out.println(rs.getString(1) + "," + rs.getString(3));
+      }
+    } finally {
+      if (rs != null) rs.close();
+      if (stmt != null) stmt.close();
+      if (conn != null) conn.close();
+    }
+  }
+}
+

+

+ + +

+ +