读写分离

语法说明

CREATE READWRITE_SPLITTING RULE readwriteSplittingRuleDefinition [, readwriteSplittingRuleDefinition] ...

ALTER READWRITE_SPLITTING RULE readwriteSplittingRuleDefinition [, readwriteSplittingRuleDefinition] ...

DROP READWRITE_SPLITTING RULE ruleName [, ruleName] ...

readwriteSplittingRuleDefinition:
    ruleName ([staticReadwriteSplittingRuleDefinition | dynamicReadwriteSplittingRuleDefinition] 
              [, loadBanlancerDefinition])

staticReadwriteSplittingRuleDefinition:
    WRITE_RESOURCE=writeResourceName, READ_RESOURCES(resourceName [, resourceName] ... )

dynamicReadwriteSplittingRuleDefinition:
    AUTO_AWARE_RESOURCE=resourceName

loadBanlancerDefinition:
    TYPE(NAME=loadBanlancerType [, PROPERTIES([algorithmProperties] )] )

algorithmProperties:
    algorithmProperty [, algorithmProperty] ...

algorithmProperty:
    key=value
  • 支持创建静态读写分离规则和动态读写分离规则
  • 动态读写分离规则依赖于数据库发现规则
  • loadBanlancerType 指定负载均衡算法类型，请参考 负载均衡算法
  • 重复的 ruleName 将无法被创建

示例

// Static
CREATE READWRITE_SPLITTING RULE ms_group_0 (
WRITE_RESOURCE=write_ds,
READ_RESOURCES(read_ds_0,read_ds_1),
TYPE(NAME=random)
);

// Dynamic
CREATE READWRITE_SPLITTING RULE ms_group_1 (
AUTO_AWARE_RESOURCE=group_0,
TYPE(NAME=random,PROPERTIES(read_weight='2:1'))
);

ALTER READWRITE_SPLITTING RULE ms_group_1 (
WRITE_RESOURCE=write_ds,
READ_RESOURCES(read_ds_0,read_ds_1,read_ds_2),
TYPE(NAME=random,PROPERTIES(read_weight='2:0'))
);

DROP READWRITE_SPLITTING RULE ms_group_1;