Influxdb cost metrics per bucket

juha-ylikoski · May 27, 2024, 8:18am

We are using Influxdb cloud 2 for our internal stuff and are looking into expanding our use of it. I’m currently trying to understand the pricing of the service and would like to be able to get per bucket statistics about the pricing vectors (InfluxDB Cloud Serverless plans | InfluxDB Cloud Serverless Documentation).

I found this article (TL;DR InfluxDB Tech Tips — Using and Understanding the InfluxDB Cloud Usage Template | InfluxData) and tried the template and was able to get:

Global data in
Global data out
Global query count
Per bucket storage

Is it somehow possible to group the data in, data out and query count with buckets so that I could see which bucket is making queries / which one is expensive?

I tried to look into the flux behind it, but the data output for the experimental apis does not seem to have buckets.

Anaisdg · May 28, 2024, 5:57pm

Hello @juha-ylikoski,
Welcome!
Where are you running Cloud 2? Through the AWS partnership?
Cloud 2 is no longer offered by Influxdata.
That dashboard does group things by bucket:

github.com

influxdata/community-templates/blob/0ab42e3c1c01e070135f5af1786f93b65cf6f52e/usage_dashboard/usage_dashboard.yml#L381


      
              type: scale
          geom: line
          height: 3
          hoverDimension: auto
          kind: Xy
          legendColorizeRows: true
          legendOpacity: 1
          legendOrientationThreshold: 1e+08
          name: Storage (By Bucket)
          position: overlaid
          queries:
            - query: "import \"experimental/usage\"\n\n//Data returned from usage.from()
                  uses the ID value for the buckets. This function will remap the data
                  to use the name of the bucket\nb = buckets()\nbucketNameForID = (id)
                  => {\n    row = b \n        |> filter(fn: (r) => r.id == id) \n        |>
                  findRecord(fn: (key) => true, idx: 0)\n    return row.name\n}\n\nbucket_ids
                  = b \n    |> findColumn(fn: (key) => true, column: \"id\")\n\ndata_buckets
                  = usage.from(start: v.timeRangeStart,\n    stop: v.timeRangeStop)
                  \n    |> filter(fn: (r) => r._measurement == \"storage_usage_bucket_bytes\")\n
                  \   |> map(fn: (r) => ({ r with bucket_name: if contains(value: r.bucket_id,
                  set: bucket_ids) then bucketNameForID(id: r.bucket_id) else \"unknown

Anaisdg · May 28, 2024, 5:58pm

@juha-ylikoski
This function also returns data for each bucket based on bucket ID which is what the template heavily uses:

juha-ylikoski · May 30, 2024, 7:32am

Hi, we have an instance in westeurope-1.azure.cloud2.influxdata.com.

Yes the dashboard in the linked article has storage by bucket but it does not have: data in, data out and query count by bucket and the usage.from() function documentation states that the http_request and query_count measurements do not have bucket names or ids.

Our original problem was that we want to know how much each team/customer costs us. Our initial plan was to have a bucket per team/customer, and we could figure out on bucket level our rough costs. Is there some other advised way to do this?

Topic		Replies	Views
Data usage per measurment / endpoint InfluxDB 2 influxdata , query	2	427	June 17, 2022
Costs of InfluxDB Cloud rises InfluxDB 2 grafana , influxdb-cloud-2-0	2	394	July 14, 2022
InfluxDB Cloud 2.0 Is Now Generally Available InfluxDB 2	3	556	September 13, 2019
Dump data from InfluxDB cloud serverless	1	174	April 3, 2024
Get data usage for token [InfluxDB 2.X OSS] InfluxDB 2	3	427	April 4, 2022

Influxdb cost metrics per bucket

Related topics