写在前面
最近一段时间都在做druid实时数据查询的工作，本文简单将官网上的英文文档加上自己的理解翻译成中文，同时将自己遇到的问题及解决方法list下，防止遗忘。
本文的demo示例均来源于官网。
Druid查询概述
Druid的查询是使用Rest风格的http请求查询服务节点，客户端通过发送Json对象请求查询接口。可以使用shell脚本查询或通过Google的ARC插件构造Post请求进行查询。
Query构建
Shell脚本
curl -X POST '<queryable_host>:<port>/druid/v2/?pretty' -H 'Content-Type:application/json' -d @<query_json_file>
其中<queryable_host>:<port>为broker、historical或realtime进程所在机器的ip和提供服务的端口，query_json_file为json配置文件路径。
ARC插件
见下图
Druid查询
1 Druid查询类型
不同的查询场景使用不同的查询方式。Druid有很多查询类型，对于各种类型的查询类型的配置可以通过配置不同的Query实现。Druid的查询类型，概括为以下3类：
1.聚合查询：时间序列查询(Timeseroes),Top查询(TopN),GroupBy查询(GroupBy) 2.元数据查询:时间范围(Time Boundary),段元数据(Segment Metadata)，数据源(DataSource) 2.Search查询(Search)
一般聚合查询使用的较多，其他类型的查询方式使用场景较少且较简单，可直接参考官网给出的demo即可查询；本文主要介绍聚合查询。一般情况下，Timeseries和TopN查询性能优于GroupBy，GroupBy查询方式最灵活但是最耗性能。Timeseries查询性能明显优于GroupBy，因为聚合不需要其他GroupBy其他维度；对于Groupby和排序在一个单一维度的场景，TopN优于GroupBy。
2 Druid主要查询属性简介
一条Druid query中主要包含以下几种属性：
1.queryType:查询类型，即timeseries，topN，groupBy等；
2.dataSource:数据源，类似Mysql中的表的概念；
3.granularity:聚合粒度，聚合粒度有none,all,week,day,hour等；
4.filter:过滤条件，类似Mysql中的where条件；
5.aggregator：聚合方式，类似Mysql中的count，sum等操作2.1 granularity简介
2.1.1 简单的聚合粒度
简单的聚合粒度有：all、none、second、minute、fifteen_minute、thirty_minute、hour、day、week、month、quarter、year；简单聚合粒度的查询取决于druid存储数据的最小粒度，如果构建数据的最小粒度是小时，使用minute粒度去查询，结果数据也是小时粒度的数据。
假设存储在Druid中的数据使用毫秒粒度构建，数据格式如下：
{"timestamp": "2013-08-31T01:02:33Z", "page": "AAA", "language" : "en"}
{"timestamp": "2013-09-01T01:02:33Z", "page": "BBB", "language" : "en"}
{"timestamp": "2013-09-02T23:32:45Z", "page": "CCC", "language" : "en"}
{"timestamp": "2013-09-03T03:32:45Z", "page": "DDD", "language" : "en"}
提交一个小时粒度的groupBy查询，查询query如下：
{
"queryType":"groupBy",
"dataSource":"my_dataSource",
"granularity":"hour",
"dimensions":[
"language"
],
"aggregations":[
{
"type":"count",
"name":"count"
}
],
"intervals":[
"2000-01-01T00:00Z/3000-01-01T00:00Z"
]
}按小时粒度进行的groupby查询结果中timestamp值精确到小时，比小时粒度更小粒度值自动补填零，以此类推按天查询，则小时及小粒度补零。timestamp值为UTC。查询结果如下：
[ {
"version" : "v1",
"timestamp" : "2013-08-31T01:00:00.000Z",
"event" : {
"count" : 1,
"language" : "en"
}
}, {
"version" : "v1",
"timestamp" : "2013-09-01T01:00:00.000Z",
"event" : {
"count" : 1,
"language" : "en"
}
}, {
"version" : "v1",
"timestamp" : "2013-09-02T23:00:00.000Z",
"event" : {
"count" : 1,
"language" : "en"
}
}, {
"version" : "v1",
"timestamp" : "2013-09-03T03:00:00.000Z",
"event" : {
"count" : 1,
"language" : "en"
}
} ]
如若指定聚合粒度为day，则按照天为单位对数据进行聚合，查询结果如下：
[ {
"version" : "v1",
"timestamp" : "2013-08-31T00:00:00.000Z",
"event" : {
"count" : 1,
"language" : "en"
}
}, {
"version" : "v1",
"timestamp" : "2013-09-01T00:00:00.000Z",
"event" : {
"count" : 1,
"language" : "en"
}
}, {
"version" : "v1",
"timestamp" : "2013-09-02T00:00:00.000Z",
"event" : {
"count" : 1,
"language" : "en"
}
}, {
"version" : "v1",
"timestamp" : "2013-09-03T00:00:00.000Z",
"event" : {
"count" : 1,
"language" : "en"
}
} ]如若聚合粒度设置为none，则按照druid中build数据的最小粒度查询数据，即不进行聚合，如bulid数据的粒度是ms，则聚合出来的结果也是毫秒：
[ {
"version" : "v1",
"timestamp" : "2013-08-31T01:02:33.000Z",
"event" : {
"count" : 1,
"language" : "en"
}
}, {
"version" : "v1",
"timestamp" : "2013-09-01T01:02:33.000Z",
"event" : {
"count" : 1,
"language" : "en"
}
}, {
"version" : "v1",
"timestamp" : "2013-09-02T23:32:45.000Z",
"event" : {
"count" : 1,
"language" : "en"
}
}, {
"version" : "v1",
"timestamp" : "2013-09-03T03:32:45.000Z",
"event" : {
"count" : 1,
"language" : "en"
}
} ]如若将聚合粒度设置为all，则返回数据的长度为1，即把查询时间段的数据做一个汇总：
[ {
"version" : "v1",
"timestamp" : "2000-01-01T00:00:00.000Z",
"event" : {
"count" : 4,
"language" : "en"
}
} ]2.1.2 时间聚合粒度
可指定一定的时间段进行聚合，返回UTC时间；支持可选属性origin；不指定时间，默认的开始时间=1970-01-01T00:00:00Z；
持续时间段2小时，从1970-01-01T00:00:00开始：
{"type": "duration", "duration": 7200000}2.1.3 常用时间段聚合粒度
时间聚合粒度的特例，方便使用，如年、月、日、小时等,日期标准是ISO 8601。无特别指定的情况下，year从1月份开始，month从1号开始，week从周一开始。
一般的格式为：其中timeZone可选，默认值是UTC；origin可选，默认1970-01-01T00:00:00；
{"type": "period", "period": "P2D", "timeZone": "America/Los_Angeles"}period的一般写法为：
month:P2M代表2个月作为一个聚合粒度；
week:P2W代表2周作为一个聚合粒度；
day:P1D代表1天作为一个聚合粒度；
hour:PT1H代表1个小时作为一个聚合粒度；
minute:PT0.750S代表750s作为一个聚合粒度；如提交一个1d作为聚合粒度的groupby查询的query：
{
"queryType":"groupBy",
"dataSource":"my_dataSource",
"granularity":{"type": "period", "period": "P1D", "timeZone": "America/Los_Angeles"},
"dimensions":[
"language"
],
"aggregations":[
{
"type":"count",
"name":"count"
}
],
"intervals":[
"1999-12-31T16:00:00.000-08:00/2999-12-31T16:00:00.000-08:00"
]
}查询得到的结果为：
[ {
"version" : "v1",
"timestamp" : "2013-08-30T00:00:00.000-07:00",
"event" : {
"count" : 1,
"language" : "en"
}
}, {
"version" : "v1",
"timestamp" : "2013-08-31T00:00:00.000-07:00",
"event" : {
"count" : 1,
"language" : "en"
}
}, {
"version" : "v1",
"timestamp" : "2013-09-02T00:00:00.000-07:00",
"event" : {
"count" : 2,
"language" : "en"
}
} ]官网给出的例子是以美国洛杉矶的时区为准，一般中国的时区这样使用，更多时区可移步该链接查询：
"granularity": {
"period": "PT1H",
"timeZone": "+08:00",
"type": "period"
}2.2 filter简介
一个filter即一个json对象，代表一个过滤条件，等价于mysql中的一个where条件；过滤器的类型主要有：Selector filter，Regular expression filter（正则表达式过滤）、Logical expression filters（AND、OR、NOT）、In filter、Bound filter、Search filter、JavaScript filter、Extraction filter；
2.2.1 Selector 过滤器
等价于 WHERE <dimension_string> = '<dimension_value_string>'
json格式：
"filter": { "type": "selector", "dimension": <dimension_string>, "value": <dimension_value_string> }2.2.2 正则表达式 过滤器
类似Selector过滤器，只不过过滤使用的是正则表达式；正则表达式为标准的java正则表达式规范；
"filter": { "type": "regex", "dimension": <dimension_string>, "pattern": <pattern_string> }2.2.3 逻辑表达式 过滤器
AND
"filter": { "type": "and", "fields": [<filter>, <filter>, ...] }OR
"filter": { "type": "not", "field": <filter> }NOT
"filter": { "type": "not", "field": <filter> }IN
等价于
SELECT COUNT(*) AS 'Count' FROM `table` WHERE `outlaw` IN ('Good', 'Bad', 'Ugly')
{
"type": "in",
"dimension": "outlaw",
"values": ["Good", "Bad", "Ugly"]
}BOUND
数值型：21<=age<=31
{ "type": "bound", "dimension": "age", "lower": "21", "upper": "31" , "ordering": "numeric" }数值型：21<age<31
{ "type": "bound", "dimension": "age", "lower": "21", "lowerStrict": true, "upper": "31" , "upperStrict": true, "ordering": "numeric" }字符型：‘foo’<=name<='hoo'
{ "type": "bound", "dimension": "name", "lower": "foo", "upper": "hoo" }2.3 aggregations简介
aggregations即汇总数据记性druid之前提供的一个数据采集一种聚合方式。常用的聚合类型主要有：count，sum，min/max，approximate，miscellaneous；
2.3.1 Count aggregator
符合查询条件的行数，类似mysql中的count计算：
{ "type" : "count", "name" : <output_name> }Note: Druid进行Count查询的数据量并不一定等于数据采集时导入的数据量，因为Druid在采集数据查询时已经按照相应的聚合方式对数据进行了聚合。
2.3.2 Sum aggregator
与底层druid表中的字段类型一致。
longSum
{ "type" : "longSum", "name" : <output_name>, "fieldName" : <metric_name> }doubleSum
{ "type" : "doubleSum", "name" : <output_name>, "fieldName" : <metric_name> }2.3.3 MIN/MAX aggregator
doubleMin
{ "type" : "doubleMin", "name" : <output_name>, "fieldName" : <metric_name> }doubleMax
{ "type" : "doubleMax", "name" : <output_name>, "fieldName" : <metric_name> }long类型类似，不在赘述。其他聚合方式请移步到官网查询示例。
2.4 聚合查询
2.4.1 Timeseries query
query
{
"queryType": "timeseries",
"dataSource": "sample_datasource",
"granularity": "day",
"descending": "true",
"filter": {
"type": "and",
"fields": [
{ "type": "selector", "dimension": "sample_dimension1", "value": "sample_value1" },
{ "type": "or",
"fields": [
{ "type": "selector", "dimension": "sample_dimension2", "value": "sample_value2" },
{ "type": "selector", "dimension": "sample_dimension3", "value": "sample_value3" }
]
}
]
},
"aggregations": [
{ "type": "longSum", "name": "sample_name1", "fieldName": "sample_fieldName1" },
{ "type": "doubleSum", "name": "sample_name2", "fieldName": "sample_fieldName2" }
],
"postAggregations": [
{ "type": "arithmetic",
"name": "sample_divide",
"fn": "/",
"fields": [
{ "type": "fieldAccess", "name": "postAgg__sample_name1", "fieldName": "sample_name1" },
{ "type": "fieldAccess", "name": "postAgg__sample_name2", "fieldName": "sample_name2" }
]
}
],
"intervals": [ "2012-01-01T00:00:00.000/2012-01-03T00:00:00.000" ]
}query属性说明
上述query的返回结果：
[
{
"timestamp": "2012-01-01T00:00:00.000Z",
"result": { "sample_name1": <some_value>, "sample_name2": <some_value>, "sample_divide": <some_value> }
},
{
"timestamp": "2012-01-02T00:00:00.000Z",
"result": { "sample_name1": <some_value>, "sample_name2": <some_value>, "sample_divide": <some_value> }
}
]2.4.2 TopN query
TopN查询根据规范返回给定维度的有序的结果集，从概念上来讲，TopN查询被认为单维度、有序的类似分组查询。在某些情况下，TopN查询比分组查询（groupby query）快。TopN查询结果返回Json数组对象。TopN在每个节点将顶上K个结果排名，在Druid默认情况下最大值为1000。在实践中，如果你要求前1000个项顺序排名，那么从第1-999个项的顺序正确性是100%，其后项的结果顺序没有保证。你可以通过增加threshold值来保证顺序准确。
query
{
"queryType": "topN",
"dataSource": "sample_data",
"dimension": "sample_dim",
"threshold": 5,
"metric": "count",
"granularity": "all",
"filter": {
"type": "and",
"fields": [
{
"type": "selector",
"dimension": "dim1",
"value": "some_value"
},
{
"type": "selector",
"dimension": "dim2",
"value": "some_other_val"
}
]
},
"aggregations": [
{
"type": "longSum",
"name": "count",
"fieldName": "count"
},
{
"type": "doubleSum",
"name": "some_metric",
"fieldName": "some_metric"
}
],
"postAggregations": [
{
"type": "arithmetic",
"name": "sample_divide",
"fn": "/",
"fields": [
{
"type": "fieldAccess",
"name": "some_metric",
"fieldName": "some_metric"
},
{
"type": "fieldAccess",
"name": "count",
"fieldName": "count"
}
]
}
],
"intervals": [
"2013-08-31T00:00:00.000/2013-09-03T00:00:00.000"
]
}query属性说明
上述query的查询结果形如：
[
{
"timestamp": "2013-08-31T00:00:00.000Z",
"result": [
{
"dim1": "dim1_val",
"count": 111,
"some_metrics": 10669,
"average": 96.11711711711712
},
{
"dim1": "another_dim1_val",
"count": 88,
"some_metrics": 28344,
"average": 322.09090909090907
},
{
"dim1": "dim1_val3",
"count": 70,
"some_metrics": 871,
"average": 12.442857142857143
},
{
"dim1": "dim1_val4",
"count": 62,
"some_metrics": 815,
"average": 13.14516129032258
},
{
"dim1": "dim1_val5",
"count": 60,
"some_metrics": 2787,
"average": 46.45
}
]
}
]2.4.3 GroupBy query
类似mysql中的groupBy查询方式。
query
{
"queryType": "topN",
"dataSource": "sample_data",
"dimension": "sample_dim",
"threshold": 5,
"metric": "count",
"granularity": "all",
"filter": {
"type": "and",
"fields": [
{
"type": "selector",
"dimension": "dim1",
"value": "some_value"
},
{
"type": "selector",
"dimension": "dim2",
"value": "some_other_val"
}
]
},
"aggregations": [
{
"type": "longSum",
"name": "count",
"fieldName": "count"
},
{
"type": "doubleSum",
"name": "some_metric",
"fieldName": "some_metric"
}
],
"postAggregations": [
{
"type": "arithmetic",
"name": "sample_divide",
"fn": "/",
"fields": [
{
"type": "fieldAccess",
"name": "some_metric",
"fieldName": "some_metric"
},
{
"type": "fieldAccess",
"name": "count",
"fieldName": "count"
}
]
}
],
"intervals": [
"2013-08-31T00:00:00.000/2013-09-03T00:00:00.000"
]
}query属性说明
上述query的查询结果形如：
[
{
"timestamp": "2013-08-31T00:00:00.000Z",
"result": [
{
"dim1": "dim1_val",
"count": 111,
"some_metrics": 10669,
"average": 96.11711711711712
},
{
"dim1": "another_dim1_val",
"count": 88,
"some_metrics": 28344,
"average": 322.09090909090907
},
{
"dim1": "dim1_val3",
"count": 70,
"some_metrics": 871,
"average": 12.442857142857143
},
{
"dim1": "dim1_val4",
"count": 62,
"some_metrics": 815,
"average": 13.14516129032258
},
{
"dim1": "dim1_val5",
"count": 60,
"some_metrics": 2787,
"average": 46.45
}
]
}
]groupBy多值字段
Druid中的字段会有多值查询，针对多值查询的groupBy操作，满足多值中一个过滤条件，查询结果中会把多值字段中的每个值都返回。下面通过例子进行说明。
底层数据格式
{"timestamp": "2011-01-12T00:00:00.000Z", "tags": ["t1","t2","t3"]} #row1
{"timestamp": "2011-01-13T00:00:00.000Z", "tags": ["t3","t4","t5"]} #row2
{"timestamp": "2011-01-14T00:00:00.000Z", "tags": ["t5","t6","t7"]} #row3
{"timestamp": "2011-01-14T00:00:00.000Z", "tags": []} #row4
查询query：
{
"queryType": "groupBy",
"dataSource": "test",
"intervals": [
"1970-01-01T00:00:00.000Z/3000-01-01T00:00:00.000Z"
],
"filter": {
"type": "selector",
"dimension": "tags",
"value": "t3"
},
"granularity": {
"type": "all"
},
"dimensions": [
{
"type": "default",
"dimension": "tags",
"outputName": "tags"
}
],
"aggregations": [
{
"type": "count",
"name": "count"
}
]
}
返回结果：命中row1和row2
[
{
"timestamp": "1970-01-01T00:00:00.000Z",
"event": {
"count": 1,
"tags": "t1"
}
},
{
"timestamp": "1970-01-01T00:00:00.000Z",
"event": {
"count": 1,
"tags": "t2"
}
},
{
"timestamp": "1970-01-01T00:00:00.000Z",
"event": {
"count": 2,
"tags": "t3"
}
},
{
"timestamp": "1970-01-01T00:00:00.000Z",
"event": {
"count": 1,
"tags": "t4"
}
},
{
"timestamp": "1970-01-01T00:00:00.000Z",
"event": {
"count": 1,
"tags": "t5"
}
}
]遇到的问题及解决方法
问题1：北京时区进行查询
解决方法：通过设置timeZone的时区解决
"granularity": {
"period": "PT1H",
"timeZone": "+08:00",
"type": "period"
}问题2：获取datasource的最新build时间
解决方法：在context中设置requireMessageWatermark=true，在http返回结果的header中拿到该数据；
if (HttpStatus.SC_OK == httpResponse.getStatusLine().getStatusCode()) {
HttpEntity entity = httpResponse.getEntity();
result.setResultStr(EntityUtils.toString(entity, "utf-8"));
String waterMark = httpResponse.getHeaders(DRUID_WATERMARK_HEADER)[0].getValue();
if (StringUtil.isNotEmpty(waterMark)) {
JSONObject obj = JSONObject.parseObject(waterMark);
waterMark = obj.getString(DRUID_MESSAGE_WATERMARK);
result.setDruidWatermarkStr(waterMark);
} // end if
}
问题3：设置不同query的优先级
解决方法：在context中设置priority属性；
问题4：设置query的超时时间
解决方法：在context中设置timeout属性；
问题5：多值groupBy指定返回想要的值
解决方法：使用listField属性设置多值列中想要的值；
{
"queryType": "groupBy",
"dataSource": "test",
"intervals": [
"1970-01-01T00:00:00.000Z/3000-01-01T00:00:00.000Z"
],
"filter": {
"type": "selector",
"dimension": "tags",
"value": "t3"
},
"granularity": {
"type": "all"
},
"dimensions": [
{
"type": "listFiltered",
"delegate": {
"type": "default",
"dimension": "tags",
"outputName": "tags"
},
"values": ["t3"]
}
],
"aggregations": [
{
"type": "count",
"name": "count"
}
]
}