vincent-zeng / beecp Goto Github PK

English|中文

BeeCP is a high-performance JDBC connection pool

Java7 or higher

<dependency>
   <groupId>com.github.chris2018998</groupId>
   <artifactId>beecp</artifactId>
   <version>3.3.0</version>
</dependency>

Java6

<dependency>
   <groupId>com.github.chris2018998</groupId>
   <artifactId>beecp</artifactId>
   <version>1.6.10</version>
</dependency>

1：Single connection threadLocal cache to improve pooling performance

2：Borrower non move waiting, saving queue entry and exit costs

3: Transfer queue reuse, which can transfer connections and exceptions

4：Connection asynchronized-add thread,which can be triggerred by releaser or waiter

Which defects of HikariCP?

1：Closed preparedStatements can be activation, when using MySQ-driver

2：When database down or network failed, getConnection response time == 'connectionTimeout'(if configed value is large,what happen?)

3：Exists transaction leak,when using 'setSavepoint' on connection

.....

Conclusion: faster, simpler, reliabler

Its usage is roughly similar to other pool, two reference examples below

BeeDataSourceConfig config = new BeeDataSourceConfig();
config.setDriverClassName("com.mysql.jdbc.Driver");
config.setJdbcUrl("jdbc:mysql://localhost/test");
config.setUsername("root");
config.setPassword("root");
BeeDataSource ds=new BeeDataSource(config);
Connection con=ds.getConnection();
....

application.properties

spring.datasource.username=root
spring.datasource.password=root
spring.datasource.url=jdbc:mysql://localhost/test
spring.datasource.driverClassName=com.mysql.jdbc.Driver

DataSourceConfig.java

@Configuration
public class DataSourceConfig {
  @Value("${spring.datasource.username}")
  private String user;
  @Value("${spring.datasource.password}")
  private String password;
  @Value("${spring.datasource.url}")
  private String url;
  @Value("${spring.datasource.driverClassName}")
  private String driver;

  @Bean
  @Primary
  @ConfigurationProperties(prefix="spring.datasource")
  public DataSource primaryDataSource() {
    return DataSourceBuilder.create().type(cn.beecp.BeeDataSource.class).build();
  }
}

☀️ If your projects are based on springboot, we recommend BeeCP-Starter to manage your datasource(file configuration, less code, monitor-ui)

If not configured, auto generated

Boolean indicator,if true,pool will use fair semaphore and fair transfer policy. default value: false

Size of connections on pool starting,if zero,pool will try to create one.default value: 0

Max reach size of connections in pool.default value: 10

Size of semaphore in pool. default value: number of CPU cores

Value setting on conneciton creating and return, default value: true

Value setting on conneciton creating and return. default value: -999,if not set then read value from first connection

Max wait time for one connection for borrower using 'getConnection'. unit: milliseconds, default value: 8000

Max idle time of connections in pool,when reach,then remove from pool.unit: milliseconds, default value: 18000

Max no-use time of borrowed connections,when reach,then return them to pool by forced close.unit: milliseconds, default value: 18000

Connection valid test sql on borrowed. default value: SELECT 1

Max time to get a valid test result. unit:second, default value: 3

Conenction valid assume time after last activity,if borrowed,not need test during the duration.unit: milliseconds, default value: 500

👉 More configuration items