Skip to content

Instantly share code, notes, and snippets.

@bartosz25
Created October 9, 2018 11:50
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save bartosz25/d0f9c4fb852a63e75f446a51dccb65e3 to your computer and use it in GitHub Desktop.
Save bartosz25/d0f9c4fb852a63e75f446a51dccb65e3 to your computer and use it in GitHub Desktop.
Apache Cassandra wide rows vs skinny rows micro-benchmark
package com.waitingforcode;
import com.datastax.driver.core.Cluster;
import com.datastax.driver.core.Session;
import java.util.ArrayList;
import java.util.List;
import java.util.stream.Collectors;
public class CassandraSession {
private static final String TEST_KEYSPACE = "test_keyspace"+System.currentTimeMillis();
private final Session session;
private final List<String> wideRowTable80Columns = new ArrayList<>();
private final List<String> wideRowTable40Columns = new ArrayList<>();
private final List<String> wideRowTable20Columns = new ArrayList<>();
public CassandraSession() {
Cluster cluster = Cluster.builder().addContactPoint("127.0.0.1").build();
session = cluster.connect();
for (int i = 0; i < 80; i++) {
String columnName = "column"+i;
wideRowTable80Columns.add(columnName);
if (i < 40) {
wideRowTable40Columns.add(columnName);
if (i < 20) {
wideRowTable20Columns.add(columnName);
}
}
}
session.execute("CREATE KEYSPACE IF NOT EXISTS "+ TEST_KEYSPACE +" WITH REPLICATION = { 'class' : 'SimpleStrategy', 'replication_factor' : 2}");
session.execute("USE "+ TEST_KEYSPACE);
}
public List<String> getWideRowTable80Columns() {
return wideRowTable80Columns;
}
public List<String> getWideRowTable40Columns() {
return wideRowTable40Columns;
}
public List<String> getWideRowTable20Columns() {
return wideRowTable20Columns;
}
public void execute(String query) {
session.execute(query);
}
public void createTables() {
session.execute(getWideRowTableCreationQuery(wideRowTable80Columns));
session.execute(getWideRowTableCreationQuery(wideRowTable40Columns));
session.execute(getWideRowTableCreationQuery(wideRowTable20Columns));
String skinnyRowTableQuery = "CREATE TABLE skinny_row_test ( col text, value text, PRIMARY KEY (col)) ";
session.execute(skinnyRowTableQuery);
}
private String getWideRowTableCreationQuery(List<String> columns) {
int columnsCount = columns.size();
String wideRowTableQuery = "CREATE TABLE wide_rows_test_"+columnsCount+" (id INT, ";
String wideRowTableCols = columns.stream().map(column -> column+" text ")
.collect(Collectors.joining(","));
wideRowTableQuery += wideRowTableCols;
wideRowTableQuery += " , PRIMARY KEY(id) )";
return wideRowTableQuery;
}
public void closeSession() {
session.close();
session.getCluster().close();
}
}
<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<modelVersion>4.0.0</modelVersion>
<groupId>com.waitingforcode</groupId>
<artifactId>wide-rows</artifactId>
<version>1.0-SNAPSHOT</version>
<name>wide-rows</name>
<properties>
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
<maven.compiler.source>1.8</maven.compiler.source>
<maven.compiler.target>1.8</maven.compiler.target>
</properties>
<dependencies>
<dependency>
<groupId>org.openjdk.jmh</groupId>
<artifactId>jmh-core</artifactId>
<version>1.21</version>
</dependency>
<dependency>
<groupId>org.openjdk.jmh</groupId>
<artifactId>jmh-generator-annprocess</artifactId>
<version>1.21</version>
</dependency>
<dependency>
<groupId>com.datastax.cassandra</groupId>
<artifactId>cassandra-driver-core</artifactId>
<version>3.6.0</version>
</dependency>
<dependency>
<groupId>com.datastax.cassandra</groupId>
<artifactId>cassandra-driver-mapping</artifactId>
<version>3.6.0</version>
</dependency>
<dependency>
<groupId>org.apache.cassandra</groupId>
<artifactId>cassandra-all</artifactId>
<version>3.11.3</version>
</dependency>
<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>
<version>4.11</version>
<scope>test</scope>
</dependency>
</dependencies>
<build>
<pluginManagement>
<plugins>
<plugin>
<artifactId>maven-clean-plugin</artifactId>
<version>3.0.0</version>
</plugin>
<plugin>
<artifactId>maven-resources-plugin</artifactId>
<version>3.0.2</version>
</plugin>
<plugin>
<artifactId>maven-compiler-plugin</artifactId>
<version>3.7.0</version>
</plugin>
<plugin>
<artifactId>maven-surefire-plugin</artifactId>
<version>2.20.1</version>
</plugin>
<plugin>
<artifactId>maven-jar-plugin</artifactId>
<version>3.0.2</version>
</plugin>
<plugin>
<artifactId>maven-install-plugin</artifactId>
<version>2.5.2</version>
</plugin>
<plugin>
<artifactId>maven-deploy-plugin</artifactId>
<version>2.8.2</version>
</plugin>
</plugins>
</pluginManagement>
</build>
</project>
package com.waitingforcode;
import org.openjdk.jmh.annotations.*;
import java.util.List;
import java.util.concurrent.TimeUnit;
import java.util.stream.Collectors;
@OutputTimeUnit(TimeUnit.MILLISECONDS)
@BenchmarkMode(Mode.All)
public class ReadMicroBenchmark {
@State(Scope.Thread)
public static class InitializationState {
private CassandraSession cassandraSession = new CassandraSession();
{
String colsToInsertInto80ColsTable = getColumnsToInsertClause(cassandraSession.getWideRowTable80Columns());
String valuesToInsert80ColsTable = getValuesToInsertClause(cassandraSession.getWideRowTable80Columns());
String colsToInsertInto40ColsTable = getColumnsToInsertClause(cassandraSession.getWideRowTable40Columns());
String valuesToInsert40ColsTable = getValuesToInsertClause(cassandraSession.getWideRowTable40Columns());
String colsToInsertInto20ColsTable = getColumnsToInsertClause(cassandraSession.getWideRowTable20Columns());
String valuesToInsert20ColsTable = getValuesToInsertClause(cassandraSession.getWideRowTable20Columns());
cassandraSession.createTables();
for (int i = 0; i < ROWS_TO_INSERT; i++) {
cassandraSession.execute("INSERT INTO wide_rows_test_80 (id, "+colsToInsertInto80ColsTable+") VALUES (" +
i+", "+valuesToInsert80ColsTable+")");
cassandraSession.execute("INSERT INTO wide_rows_test_40 (id, "+colsToInsertInto40ColsTable+") VALUES (" +
i+", "+valuesToInsert40ColsTable+")");
cassandraSession.execute("INSERT INTO wide_rows_test_20 (id, "+colsToInsertInto20ColsTable+") VALUES (" +
i+", "+valuesToInsert20ColsTable+")");
for (int j = 0; j < cassandraSession.getWideRowTable80Columns().size(); j++) {
String key = i+"_"+cassandraSession.getWideRowTable80Columns().get(j);
cassandraSession.execute("INSERT INTO skinny_row_test (col, value) VALUES ('"+key+"', 'value"+i+"')");
}
}
}
private static final int ROWS_TO_INSERT = 100;
private static String getColumnsToInsertClause(List<String> columns) {
return columns.stream().collect(Collectors.joining(","));
}
private static String getValuesToInsertClause(List<String> columns) {
return columns.stream()
.map(colName -> "'"+colName + " value'")
.collect(Collectors.joining(","));
}
@Setup(Level.Trial)
public void doSetup() {
}
@TearDown(Level.Trial)
public void doTearDown() {
cassandraSession.closeSession();
}
}
@Benchmark
public void read_all_wide_rows_80_cols(InitializationState initializationState) {
for (int i = 0; i < initializationState.ROWS_TO_INSERT; i++) {
initializationState.cassandraSession.execute("SELECT * FROM wide_rows_test_80 WHERE id = "+i);
}
}
@Benchmark
public void read_all_wide_rows_40_cols(InitializationState initializationState) {
for (int i = 0; i < initializationState.ROWS_TO_INSERT; i++) {
initializationState.cassandraSession.execute("SELECT * FROM wide_rows_test_40 WHERE id = "+i);
}
}
@Benchmark
public void read_all_wide_rows_20_cols(InitializationState initializationState) {
for (int i = 0; i < initializationState.ROWS_TO_INSERT; i++) {
initializationState.cassandraSession.execute("SELECT * FROM wide_rows_test_20 WHERE id = "+i);
}
}
@Benchmark
public void read_all_skinny_rows(InitializationState initializationState) {
for (int i = 0; i < initializationState.ROWS_TO_INSERT; i++) {
for (int j = 0; j < initializationState.cassandraSession.getWideRowTable80Columns().size(); j++) {
String key = i+"_"+initializationState.cassandraSession.getWideRowTable80Columns().get(j);
initializationState.cassandraSession.execute("SELECT * FROM skinny_row_test WHERE col = '"+key+"'");
}
}
}
@Benchmark
public void read_20_columns_from_all_wide_row_80_cols_table(InitializationState initializationState) {
for (int i = 0; i < initializationState.ROWS_TO_INSERT; i++) {
initializationState.cassandraSession.execute("SELECT id, column0, column1, column2, column3, column4, column5, column6, column7, column8," +
"column9, column10, column11, column12, column13, column14, column15, column16, column17, " +
"column18, column19 FROM wide_rows_test_80 WHERE id = "+i);
}
}
@Benchmark
public void read_20_columns_from_all_wide_row_40_cols_table(InitializationState initializationState) {
for (int i = 0; i < initializationState.ROWS_TO_INSERT; i++) {
initializationState.cassandraSession.execute("SELECT id, column0, column1, column2, column3, column4, column5, column6, column7, column8," +
"column9, column10, column11, column12, column13, column14, column15, column16, column17, " +
"column18, column19 FROM wide_rows_test_40 WHERE id = "+i);
}
}
@Benchmark
public void read_20_columns_from_all_wide_row_20_cols_table(InitializationState initializationState) {
for (int i = 0; i < initializationState.ROWS_TO_INSERT; i++) {
initializationState.cassandraSession.execute("SELECT id, column0, column1, column2, column3, column4, column5, column6, column7, column8," +
"column9, column10, column11, column12, column13, column14, column15, column16, column17, " +
"column18, column19 FROM wide_rows_test_20 WHERE id = "+i);
}
}
@Benchmark
public void read_20_rows_from_all_skinny_rows(InitializationState initializationState) {
for (int i = 0; i < initializationState.ROWS_TO_INSERT; i++) {
for (int j = 0; j < 20; j++) {
String key = i+"_"+initializationState.cassandraSession.getWideRowTable80Columns().get(j);
initializationState.cassandraSession.execute("SELECT * FROM skinny_row_test WHERE col = '"+key+"'");
}
}
}
}
package com.waitingforcode;
import org.openjdk.jmh.runner.Runner;
import org.openjdk.jmh.runner.RunnerException;
import org.openjdk.jmh.runner.options.Options;
import org.openjdk.jmh.runner.options.OptionsBuilder;
public class ReadMicroBenchmarkRunner {
public static void main(String[] args) throws RunnerException {
Options opt = new OptionsBuilder()
.include(ReadMicroBenchmark.class.getSimpleName())
.forks(1)
.threads(1)
.warmupIterations(0)
.measurementIterations(3)
.output("read_benchmark_debug.txt")
.result("read_benchmark_results.txt")
.build();
new Runner(opt).run();
}
}
package com.waitingforcode;
import org.openjdk.jmh.annotations.*;
import java.util.List;
import java.util.concurrent.TimeUnit;
import java.util.stream.Collectors;
@OutputTimeUnit(TimeUnit.MILLISECONDS)
@BenchmarkMode(Mode.All)
public class WriteMicroBenchmark {
@State(Scope.Thread)
public static class InitializationState {
private CassandraSession cassandraSession = new CassandraSession();
private String colsToInsertInto80 = getColsToInertClause(cassandraSession.getWideRowTable80Columns());
private String valuesToInsertInto80 = getValuesToInsertClause(cassandraSession.getWideRowTable80Columns());
private String colsToInsertInto40 = getColsToInertClause(cassandraSession.getWideRowTable40Columns());
private String valuesToInsertInto40 = getValuesToInsertClause(cassandraSession.getWideRowTable40Columns());
private String colsToInsertInto20 = getColsToInertClause(cassandraSession.getWideRowTable20Columns());
private String valuesToInsertInto20 = getValuesToInsertClause(cassandraSession.getWideRowTable20Columns());
private static final int ROWS_TO_INSERT = 100;
private static String getColsToInertClause(List<String> columns) {
return columns.stream().collect(Collectors.joining(","));
}
private static String getValuesToInsertClause(List<String> columns) {
return columns.stream()
.map(colName -> "'"+colName + " value'")
.collect(Collectors.joining(","));
}
@Setup(Level.Trial)
public void doSetup() {
cassandraSession.createTables();
}
@TearDown(Level.Trial)
public void doTearDown() {
cassandraSession.closeSession();
}
}
@Benchmark
public void write_wide_rows_to_80_cols_table(InitializationState initializationState) {
for (int i = 0; i < initializationState.ROWS_TO_INSERT; i++) {
initializationState.cassandraSession.execute("INSERT INTO wide_rows_test_80 (id, "+initializationState.colsToInsertInto80 +") VALUES (" +
i+", "+initializationState.valuesToInsertInto80 +")");
}
}
@Benchmark
public void write_wide_rows_to_40_cols_table(InitializationState initializationState) {
for (int i = 0; i < initializationState.ROWS_TO_INSERT; i++) {
initializationState.cassandraSession.execute("INSERT INTO wide_rows_test_40 (id, "+initializationState.colsToInsertInto40 +") VALUES (" +
i+", "+initializationState.valuesToInsertInto40 +")");
}
}
@Benchmark
public void write_wide_rows_to_20_cols_table(InitializationState initializationState) {
for (int i = 0; i < initializationState.ROWS_TO_INSERT; i++) {
initializationState.cassandraSession.execute("INSERT INTO wide_rows_test_20 (id, "+initializationState.colsToInsertInto20 +") VALUES (" +
i+", "+initializationState.valuesToInsertInto20 +")");
}
}
@Benchmark
public void write_skinny_rows(InitializationState initializationState) {
for (int i = 0; i < initializationState.ROWS_TO_INSERT; i++) {
for (int j = 0; j < initializationState.cassandraSession.getWideRowTable80Columns().size(); j++) {
String key = initializationState.cassandraSession.getWideRowTable80Columns().get(j);
initializationState.cassandraSession.execute("INSERT INTO skinny_row_test (col, value) VALUES ('"+key+"', 'value "+key+"')");
}
}
}
}
package com.waitingforcode;
import org.openjdk.jmh.runner.Runner;
import org.openjdk.jmh.runner.RunnerException;
import org.openjdk.jmh.runner.options.Options;
import org.openjdk.jmh.runner.options.OptionsBuilder;
public class WriteMicroBenchmarkRunner {
public static void main(String[] args) throws RunnerException {
Options opt = new OptionsBuilder()
.include(WriteMicroBenchmark.class.getSimpleName())
.forks(1)
.threads(1)
.warmupIterations(0)
.measurementIterations(3)
.output("write_benchmark_debug.txt")
.result("write_benchmark_results.txt")
.build();
new Runner(opt).run();
}
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment