# Aggregating min and max for many fields

**URL:** <https://community.influxdata.com/t/aggregating-min-and-max-for-many-fields/29159>\
**Category:** InfluxDB 2\
**Created:** [March 21, 2023, 8:36pm UTC](https://community.influxdata.com/t/aggregating-min-and-max-for-many-fields/29159 "2023-03-21T20:36:50Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![dondario](https://avatars.discourse-cdn.com/v4/letter/d/2acd7d/32.png) [@dondario](https://community.influxdata.com/u/dondario)\
**Post date:** [March 21, 2023, 8:36pm UTC](https://community.influxdata.com/t/aggregating-min-and-max-for-many-fields/29159/1 "2023-03-21T20:36:50Z")

</div>

Hi,

i’m Dario from Bochum, Germany and new to Influx 2.x (migrating from Influx 1).

I store data in a bucket and now, i want to aggregate that data using a task, which will called once every day.

To test it, i have raw data from one year in the test-bucket “sensors”  
and i want to aggregate them in a bucket with infinite retention called “forever”.  
This works as expected for

- mean values (4h window)
- min and max values (12h window)

with this queries:

```auto
from(bucket: "sensors")
  |> range(start: -365d)
  |> filter(fn: (r) => r["_measurement"] == "Weather")
  |> filter(fn: (r) => r["_field"] == "SolarRadiation")
  |> aggregateWindow(every: 4h, fn: mean)
  |> set(key: "_field", value: "SolarRadiation_mean")
  |> to(bucket: "forever")  

from(bucket: "sensors")
  |> range(start: -365d)
  |> filter(fn: (r) => r["_measurement"] == "Weather")
  |> filter(fn: (r) => r["_field"] == "SolarRadiation")
  |> aggregateWindow(every: 12h, fn: max)
  |> set(key: "_field", value: "SolarRadiation_max")
  |> to(bucket: "forever")  

from(bucket: "sensors")
  |> range(start: -365d)
  |> filter(fn: (r) => r["_measurement"] == "Weather")
  |> filter(fn: (r) => r["_field"] == "SolarRadiation")
  |> aggregateWindow(every: 12h, fn: min)
  |> set(key: "_field", value: "SolarRadiation_min")
  |> to(bucket: "forever")  

```

For tasks i would change “range(start: -365d)” to “range(start: -task.every)”

But when i want do this for much more fields, like:

- Temperature
- Humidity
- Pressure
- WindSpeed
- WindDirection
- GustSpeed
- Rainrate
- GoldcapVoltage
- RainClicks  
an only max-values for
- RainToday
- Longest Blackout
- Packets received
- CRC-Errors
- BattWarning

I will end up in 3\*10 + 5 = 35 Querries = 35 Tasks.

I was wondering if there is an elegant method to archive this goal.

Hoping to hear an elegant solution

Greetings from Bochum.

---

<div class="post-metadata">

**Author:** ![Anaisdg](https://sea1.discourse-cdn.com/flex023/user_avatar/community.influxdata.com/anaisdg/32/6401_2.png) [@Anaisdg](https://community.influxdata.com/u/Anaisdg)\
**Post date:** [March 21, 2023, 9:18pm UTC](https://community.influxdata.com/t/aggregating-min-and-max-for-many-fields/29159/2 "2023-03-21T21:18:38Z")

</div>

Hello @dondario,  
Do you really want to rename everything would it be okay not too? That’ll simplify the query.  
You can do:

```auto
data = from(bucket: "sensors")
  |> range(start: -365d)
  |> filter(fn: (r) => r["_measurement"] == "Weather")
  |> filter(fn: (r) => r["_field"] == "SolarRadiation" or r["_field"] == "Temperature" or r["_field"] == "Pressure" or ...)
data
  |> aggregateWindow(every: 4h, fn: mean)
  |> to(bucket: "forever")  

data
  |> aggregateWindow(every: 4h, fn: min)
  |> to(bucket: "forever")  

data
  |> aggregateWindow(every: 4h, fn: max)
  |> to(bucket: "forever")  

```

Alternatively if you are querying every field in your measurement you can just do:

```auto
data = from(bucket: "sensors")
  |> range(start: -365d)
  |> filter(fn: (r) => r["_measurement"] == "Weather")

```

Lastly if you’re applying these aggregates to every field except a couple you can alternatively do:

```auto
from(bucket: "sensors")
  |> range(start: -365d)
  |> filter(fn: (r) => r["_measurement"] == "Weather")
  |> filter(fn: (r) => r["_field"] != "only field you dont want to calculate the mean for")
  |> aggregateWindow(every: 4h, fn: mean)
  |> set(key: "_field", value: "SolarRadiation_mean")
  |> to(bucket: "forever")  

```

But yes you’ll want to apply an aggregateWindow to all of your data for each aggreagation so the general structure will be the same.

However the following might interest you:

```auto
minMaxMean = (tables=<-) =>
  tables
    |> reduce(
      identity: {count: 0.0, sum: 0.0, min: 0.0, max: 0.0, mean: 0.0},
      fn: (r, accumulator) => ({
        count: accumulator.count + 1.0,
        sum: r._value + accumulator.sum,
        min: if accumulator.count == 0.0 then r._value else if r._value < accumulator.min then r._value else accumulator.min,
        max: if accumulator.count == 0.0 then r._value else if r._value > accumulator.max then r._value else accumulator.max,
        mean: (r._value + accumulator.sum) / (accumulator.count + 1.0)
      })
    )

```

And the following posts:

> [@Combined min, max, mean and sum query - best way?](https://community.influxdata.com/t/combined-min-max-mean-and-sum-query-best-way/12570):
>
> I have a query I run in my OSS 1.7 influxdb that looks like this: SELECT mean('temperature') AS MeanTemp, max('temperature') AS MaxTemp, min('temperature') AS MinTemp, mean('flow') AS MeanFlow, max('flow') AS MaxFlow, min('flow') AS MinFlow, sum('credit') AS TotalCredits FROM MyMeasurement WHERE time \>= startTime and time \<= endTime The result is a single row that has the mean, min and max values for the temperature and flow and the total credits for the defined time window. I’d like to replic…

> [@Calculate value range of a field/column](https://community.influxdata.com/t/calculate-value-range-of-a-field-column/17357):
>
> Hello, i want to calculate the value range of a field in a time range. With InfluxQL i would simply use: SELECT max(fieldname)-min(fieldname) FROM measurement where time \> now()-120s I found no solution to do the above with flux. I tried to transpile with influx tool but only get: panic: interface conversion: \*ast.BinaryExpression is not ast.PropertyKey: missing method Key Thanks in advance for your help, Harald Tillmanns

> [@How to handle two fields in flux](https://community.influxdata.com/t/how-to-handle-two-fields-in-flux/11322/2):
>
> @Sameer_Raj You’re correct in saying that each table is its own line in the graph. So the goal is to ensure that f1 and f2 are in separate tables, which they should be by default since \_field is part of the group key. In your results, it’s important to understand that f1 and f2 are not column names. They are values in the \_field column. To get them to be their own columns, you need to use pivot() right after your filter. from(bucket: "SomeDB/autogen") |\> range($range) |\> filter(fn: (r) =\> …

But Id still use 3 separate aggwindow functions