Performance regression on closing statement with Hikari

[NathanEckert]

Driver version

Upgrade from driver 2.1.0.19 to 2.1.0.20

Redshift version

PostgreSQL 8.0.2 on i686-pc-linux-gnu, compiled by GCC gcc (GCC) 3.4.2 20041017 (Red Hat 3.4.2-6.fc3), Redshift 1.0.57864

Client Operating System

Linux Ubuntu 22.04.3 LTS (Jammy Jellyfish)

JAVA/JVM version

Apache Maven 3.8.6 (84538c9988a25aec085021c365c560670ad80f63)
Maven home: /home/nec/.sdkman/candidates/maven/current
Java version: 11.0.20, vendor: Eclipse Adoptium, runtime: /home/nec/.sdkman/candidates/java/11.0.20-tem
Default locale: en_US, platform encoding: UTF-8
OS name: "linux", version: "6.2.0-35-generic", arch: "amd64", family: "unix"

Additional dependency:

<dependency>
<groupId>com.zaxxer</groupId>
<artifactId>HikariCP</artifactId>
<version>5.0.1</version>
</dependency>

Problem description

When upgrading driver version, a performance regression is observed when using the driver in conjunction with Hikari.

Before the upgrade, it took 13 ms to close the statement, versus more than 2 seconds after the upgrade.

This issue is only observed with Hikari.

Reproduction code

package com.activeviam.database.redshift;

import com.activeviam.directquery.test.internal.redshift.RedshiftTestConstants;
import com.amazon.redshift.RedshiftProperty;
import com.zaxxer.hikari.HikariConfig;
import com.zaxxer.hikari.HikariDataSource;
import java.sql.Connection;
import java.sql.PreparedStatement;
import java.sql.SQLException;
import java.util.Properties;
import java.util.concurrent.TimeUnit;
import lombok.Cleanup;
import org.junit.jupiter.api.Test;

public class TestPerformanceJDBCDriver {

	private static final String CONNECTION_STRING = RedshiftTestConstants.REDSHIFT_CLUSTER_URL;
	private static final String QUERY = "SELECT 1;";


	@Test
	void testPreparedStatement() throws SQLException {

		Connection connection = null;
		PreparedStatement statement = null;

		@Cleanup
		final HikariDataSource dataSource = createHikariDatasource(CONNECTION_STRING + RedshiftTestConstants.TEST_DATABASE_NAME);

		try {
			connection = dataSource.getConnection();
			statement = connection.prepareStatement(QUERY);
			statement.execute();
		} catch (SQLException e) {
			throw new RuntimeException(e);
		} finally {

			if (statement != null) {
				final long startClosingStatement = System.nanoTime();
				statement.close();
				System.out.println("Closing statement took: " + TimeUnit.NANOSECONDS.toMillis(System.nanoTime() - startClosingStatement) + " ms.");
			}
			if (connection != null) {
				connection.close();
			}

		}


	}

	private static HikariDataSource createHikariDatasource(final String connectionString) {
		final Properties jdbcProperties = new Properties();
		RedshiftProperty.USER.set(jdbcProperties, RedshiftTestConstants.getUsername());
		RedshiftProperty.PASSWORD.set(jdbcProperties, RedshiftTestConstants.getRedshiftPassword());

		final HikariConfig config = new HikariConfig();
		config.setDataSourceProperties(jdbcProperties);
		config.setJdbcUrl(connectionString);
		return new HikariDataSource(config);
	}
}

The previous outputs Closing statement took: 2021 ms. with driver version 2.1.0.20, and Closing statement took: 14 ms. with driver version 2.1.0.19

[bhvkshah]

@NathanEckert we are aware of this issue and have a fix in place. It should available as part of the next release, version 2.1.0.22, slated for mid-November

[bhvkshah]

Hi @NathanEckert , driver version 2.1.0.22 has been released which contains the fix for this issue. Thank you again for your contribution to the Redshift JDBC Driver!