使用Flux窗口和聚合数据
此页面记录了 InfluxDB OSS 的早期版本。InfluxDB OSS v2 是最新的稳定版本。请参阅等效的 InfluxDB v2 文档： 使用 Flux 窗口和聚合数据。
对时间序列数据执行的一个常见操作是将数据分组为时间窗口，或称为“窗口化”数据，然后将窗口中的值聚合成一个新值。 本指南介绍了如何使用Flux进行数据的窗口化和聚合，并演示了数据在此过程中是如何被塑造的。
如果您刚刚开始使用Flux查询，请查看以下内容：
以下示例详细介绍了窗口化和聚合数据所需的步骤。 aggregateWindow() 函数 为您执行这些操作，但理解数据在这一过程中是如何被塑造的，有助于成功创建您所期望的输出。
数据集
为了本指南的目的，定义一个表示您的基础数据集的变量。 以下示例查询主机机器的内存使用情况。
dataSet = from(bucket: "db/rp")
|> range(start: -5m)
|> filter(fn: (r) =>
r._measurement == "mem" and
r._field == "used_percent"
)
|> drop(columns: ["host"])
此示例删除了返回数据中的 host 列，因为内存数据仅针对单个主机进行跟踪，这简化了输出表。删除 host 列是可选的，如果要监控多个主机的内存，则不建议这么做。
dataSet 现在可以用来表示你的基础数据，其外观将类似于以下内容：
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _time:time _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ------------------------------ ----------------------------
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:50:00.000000000Z 71.11611366271973
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:50:10.000000000Z 67.39630699157715
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:50:20.000000000Z 64.16666507720947
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:50:30.000000000Z 64.19951915740967
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:50:40.000000000Z 64.2122745513916
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:50:50.000000000Z 64.22209739685059
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:51:00.000000000Z 64.6336555480957
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:51:10.000000000Z 64.16516304016113
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:51:20.000000000Z 64.18349742889404
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:51:30.000000000Z 64.20474052429199
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:51:40.000000000Z 68.65062713623047
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:51:50.000000000Z 67.20139980316162
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:52:00.000000000Z 70.9143877029419
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:52:10.000000000Z 64.14549350738525
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:52:20.000000000Z 64.15379047393799
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:52:30.000000000Z 64.1592264175415
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:52:40.000000000Z 64.18190002441406
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:52:50.000000000Z 64.28837776184082
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:53:00.000000000Z 64.29731845855713
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:53:10.000000000Z 64.36963081359863
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:53:20.000000000Z 64.37397003173828
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:53:30.000000000Z 64.44413661956787
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:53:40.000000000Z 64.42906856536865
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:53:50.000000000Z 64.44573402404785
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:54:00.000000000Z 64.48912620544434
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:54:10.000000000Z 64.49522972106934
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:54:20.000000000Z 64.48652744293213
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:54:30.000000000Z 64.49949741363525
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:54:40.000000000Z 64.4949197769165
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:54:50.000000000Z 64.49787616729736
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:55:00.000000000Z 64.49816226959229
窗口化数据
使用window()函数根据时间范围对数据进行分组。与window()一起传递的最常见参数是every，它定义了窗口之间的时间持续时间。其他参数也可用，但对于这个示例，将基础数据集窗口化为一分钟的窗口。
dataSet
|> window(every: 1m)
每个 every 参数支持所有 有效的持续时间单位，包括 日历月份 (1mo) 和 年度 (1y)。
每个时间窗口都在其自己的表中输出，包含所有在该窗口内的记录。
window() 输出表
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _time:time _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ------------------------------ ----------------------------
2018-11-03T17:50:00.000000000Z 2018-11-03T17:51:00.000000000Z used_percent mem 2018-11-03T17:50:00.000000000Z 71.11611366271973
2018-11-03T17:50:00.000000000Z 2018-11-03T17:51:00.000000000Z used_percent mem 2018-11-03T17:50:10.000000000Z 67.39630699157715
2018-11-03T17:50:00.000000000Z 2018-11-03T17:51:00.000000000Z used_percent mem 2018-11-03T17:50:20.000000000Z 64.16666507720947
2018-11-03T17:50:00.000000000Z 2018-11-03T17:51:00.000000000Z used_percent mem 2018-11-03T17:50:30.000000000Z 64.19951915740967
2018-11-03T17:50:00.000000000Z 2018-11-03T17:51:00.000000000Z used_percent mem 2018-11-03T17:50:40.000000000Z 64.2122745513916
2018-11-03T17:50:00.000000000Z 2018-11-03T17:51:00.000000000Z used_percent mem 2018-11-03T17:50:50.000000000Z 64.22209739685059
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _time:time _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ------------------------------ ----------------------------
2018-11-03T17:51:00.000000000Z 2018-11-03T17:52:00.000000000Z used_percent mem 2018-11-03T17:51:00.000000000Z 64.6336555480957
2018-11-03T17:51:00.000000000Z 2018-11-03T17:52:00.000000000Z used_percent mem 2018-11-03T17:51:10.000000000Z 64.16516304016113
2018-11-03T17:51:00.000000000Z 2018-11-03T17:52:00.000000000Z used_percent mem 2018-11-03T17:51:20.000000000Z 64.18349742889404
2018-11-03T17:51:00.000000000Z 2018-11-03T17:52:00.000000000Z used_percent mem 2018-11-03T17:51:30.000000000Z 64.20474052429199
2018-11-03T17:51:00.000000000Z 2018-11-03T17:52:00.000000000Z used_percent mem 2018-11-03T17:51:40.000000000Z 68.65062713623047
2018-11-03T17:51:00.000000000Z 2018-11-03T17:52:00.000000000Z used_percent mem 2018-11-03T17:51:50.000000000Z 67.20139980316162
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _time:time _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ------------------------------ ----------------------------
2018-11-03T17:52:00.000000000Z 2018-11-03T17:53:00.000000000Z used_percent mem 2018-11-03T17:52:00.000000000Z 70.9143877029419
2018-11-03T17:52:00.000000000Z 2018-11-03T17:53:00.000000000Z used_percent mem 2018-11-03T17:52:10.000000000Z 64.14549350738525
2018-11-03T17:52:00.000000000Z 2018-11-03T17:53:00.000000000Z used_percent mem 2018-11-03T17:52:20.000000000Z 64.15379047393799
2018-11-03T17:52:00.000000000Z 2018-11-03T17:53:00.000000000Z used_percent mem 2018-11-03T17:52:30.000000000Z 64.1592264175415
2018-11-03T17:52:00.000000000Z 2018-11-03T17:53:00.000000000Z used_percent mem 2018-11-03T17:52:40.000000000Z 64.18190002441406
2018-11-03T17:52:00.000000000Z 2018-11-03T17:53:00.000000000Z used_percent mem 2018-11-03T17:52:50.000000000Z 64.28837776184082
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _time:time _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ------------------------------ ----------------------------
2018-11-03T17:53:00.000000000Z 2018-11-03T17:54:00.000000000Z used_percent mem 2018-11-03T17:53:00.000000000Z 64.29731845855713
2018-11-03T17:53:00.000000000Z 2018-11-03T17:54:00.000000000Z used_percent mem 2018-11-03T17:53:10.000000000Z 64.36963081359863
2018-11-03T17:53:00.000000000Z 2018-11-03T17:54:00.000000000Z used_percent mem 2018-11-03T17:53:20.000000000Z 64.37397003173828
2018-11-03T17:53:00.000000000Z 2018-11-03T17:54:00.000000000Z used_percent mem 2018-11-03T17:53:30.000000000Z 64.44413661956787
2018-11-03T17:53:00.000000000Z 2018-11-03T17:54:00.000000000Z used_percent mem 2018-11-03T17:53:40.000000000Z 64.42906856536865
2018-11-03T17:53:00.000000000Z 2018-11-03T17:54:00.000000000Z used_percent mem 2018-11-03T17:53:50.000000000Z 64.44573402404785
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _time:time _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ------------------------------ ----------------------------
2018-11-03T17:54:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:54:00.000000000Z 64.48912620544434
2018-11-03T17:54:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:54:10.000000000Z 64.49522972106934
2018-11-03T17:54:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:54:20.000000000Z 64.48652744293213
2018-11-03T17:54:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:54:30.000000000Z 64.49949741363525
2018-11-03T17:54:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:54:40.000000000Z 64.4949197769165
2018-11-03T17:54:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:54:50.000000000Z 64.49787616729736
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _time:time _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ------------------------------ ----------------------------
2018-11-03T17:55:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:55:00.000000000Z 64.49816226959229
在InfluxDB UI中可视化时，每个窗口表以不同的颜色显示。
聚合数据
聚合函数 获取表中所有行的值，并使用这些值执行聚合操作。结果作为一个新值输出到一个单行表中。
由于带窗口的数据被分割成单独的表，因此聚合操作针对每个表单独运行，并输出仅包含聚合值的新表。
在这个例子中，使用 mean() 函数 输出每个窗口的平均值：
dataSet
|> window(every: 1m)
|> mean()
mean() 输出表
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ----------------------------
2018-11-03T17:50:00.000000000Z 2018-11-03T17:51:00.000000000Z used_percent mem 65.88549613952637
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ----------------------------
2018-11-03T17:51:00.000000000Z 2018-11-03T17:52:00.000000000Z used_percent mem 65.50651391347249
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ----------------------------
2018-11-03T17:52:00.000000000Z 2018-11-03T17:53:00.000000000Z used_percent mem 65.30719598134358
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ----------------------------
2018-11-03T17:53:00.000000000Z 2018-11-03T17:54:00.000000000Z used_percent mem 64.39330975214641
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ----------------------------
2018-11-03T17:54:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 64.49386278788249
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ----------------------------
2018-11-03T17:55:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 64.49816226959229
因为每个数据点都包含在自己的表中，当可视化时，它们看起来像是单独且不相连的点。
重新创建时间列
请注意 _time 列不在 聚合输出表 中。
因为每个表中的记录被聚合在一起，它们的时间戳不再适用，因此该列从组键和表中移除。
还要注意 _start 和 _stop 列仍然存在。 这些表示时间窗口的下限和上限。
许多 Flux 函数依赖于 _time 列。
在聚合函数处理后要进一步处理数据，您需要重新添加 _time。
使用 duplicate() 函数来
将 _start 或 _stop 列复制为新的 _time 列。
dataSet
|> window(every: 1m)
|> mean()
|> duplicate(column: "_stop", as: "_time")
duplicate() 输出表
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _time:time _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ------------------------------ ----------------------------
2018-11-03T17:50:00.000000000Z 2018-11-03T17:51:00.000000000Z used_percent mem 2018-11-03T17:51:00.000000000Z 65.88549613952637
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _time:time _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ------------------------------ ----------------------------
2018-11-03T17:51:00.000000000Z 2018-11-03T17:52:00.000000000Z used_percent mem 2018-11-03T17:52:00.000000000Z 65.50651391347249
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _time:time _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ------------------------------ ----------------------------
2018-11-03T17:52:00.000000000Z 2018-11-03T17:53:00.000000000Z used_percent mem 2018-11-03T17:53:00.000000000Z 65.30719598134358
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _time:time _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ------------------------------ ----------------------------
2018-11-03T17:53:00.000000000Z 2018-11-03T17:54:00.000000000Z used_percent mem 2018-11-03T17:54:00.000000000Z 64.39330975214641
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _time:time _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ------------------------------ ----------------------------
2018-11-03T17:54:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:55:00.000000000Z 64.49386278788249
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _time:time _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ------------------------------ ----------------------------
2018-11-03T17:55:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:55:00.000000000Z 64.49816226959229
“Unwindow” 聚合表
将汇总值保存在单独的表中通常不是您想要的数据格式。使用window()函数将您的数据“去窗”到一个单一的无限 (inf) 窗口中。
dataSet
|> window(every: 1m)
|> mean()
|> duplicate(column: "_stop", as: "_time")
|> window(every: inf)
窗口处理需要一个 _time 列，这就是为什么在聚合之后有必要重新创建 _time 列。
无窗口输出表
Table: keys: [_start, _stop, _field, _measurement]
_start:time _stop:time _field:string _measurement:string _time:time _value:float
------------------------------ ------------------------------ ---------------------- ---------------------- ------------------------------ ----------------------------
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:51:00.000000000Z 65.88549613952637
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:52:00.000000000Z 65.50651391347249
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:53:00.000000000Z 65.30719598134358
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:54:00.000000000Z 64.39330975214641
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:55:00.000000000Z 64.49386278788249
2018-11-03T17:50:00.000000000Z 2018-11-03T17:55:00.000000000Z used_percent mem 2018-11-03T17:55:00.000000000Z 64.49816226959229
通过在一个表中聚合值，数据点在可视化中相互连接。
总结
您现在已经创建了一个 Flux 查询，该查询对数据进行分窗口和聚合。 本指南中概述的数据转换过程应适用于所有聚合操作。
Flux 还提供了 aggregateWindow() 函数，该函数为您执行所有这些单独的功能。
以下 Flux 查询将返回相同的结果：
聚合窗口函数
dataSet
|> aggregateWindow(every: 1m, fn: mean)