Add percentiles bucket pipeline aggregation

[joejuzl]

Describe the feature:
It would be great to be able to use the percentiles_bucket aggregation in a metrics visualisation.

Describe a specific use case for the feature:
I store events in elasticsearch. Each event has a user and a timestamp.
I have a scripted field event_delay which calculates the time difference between now and the timestamp.
I can find the min event_delay, let's call it diary_delay, for a single user easily. Using the min aggregation.
However I want to the percentiles of diary_delay aggregated over all users.
In elasticsearch this would be done like so:

{
  "size": 0,
  "aggs": {
    "diary_delay_per_user": {
      "terms": {
        "field": "user.keyword"
      },
      "aggs": {
        "diary_delay": {
          "min": {
            "script" : {
              "lang": "painless",
              "source": "(new Date().getTime() - doc['end_timestamp'].value.getMillis())/3600000"
            }
          }
        }
      }
    },
    "avg_diary_delay": {
      "avg_bucket": {
        "buckets_path": "diary_delay_per_user>diary_delay"
      }
    },
    "percentiles_diary_delay": {
      "percentiles_bucket": {
        "buckets_path": "diary_delay_per_user>diary_delay"
      }
    }
  }
}

However without percentiles_diary_delay I can't replicate this in kibana. There may be an alternative that I am not aware of.

Thanks!

[TraderApophis]

+1

[timductive]

Closing this because it's not planned to be resolved in the foreseeable future. It will be tracked in our Icebox and will be re-opened if our priorities change. Feel free to re-open if you think it should be melted sooner.