Composite aggregation

edit

The composite aggregation is expensive. Load test your application before deploying a composite aggregation in production.

A multi-bucket aggregation that creates composite buckets from different sources.

Unlike the other multi-bucket aggregations, you can use the composite aggregation to paginate all buckets from a multi-level aggregation efficiently. This aggregation provides a way to stream all buckets of a specific aggregation, similar to what scroll does for documents.

The composite buckets are built from the combinations of the values extracted/created for each document and each combination is considered as a composite bucket.

For example, consider the following document:

{
  "keyword": ["foo", "bar"],
  "number": [23, 65, 76]
}

Using keyword and number as source fields for the aggregation results in the following composite buckets:

{ "keyword": "foo", "number": 23 }
{ "keyword": "foo", "number": 65 }
{ "keyword": "foo", "number": 76 }
{ "keyword": "bar", "number": 23 }
{ "keyword": "bar", "number": 65 }
{ "keyword": "bar", "number": 76 }

Value sources

edit

The sources parameter defines the source fields to use when building composite buckets. The order that the sources are defined controls the order that the keys are returned.

You must use a unique name when defining sources.

The sources parameter can be any of the following types:

Terms

edit

The terms value source is equivalent to a simple terms aggregation. The values are extracted from a field exactly like the terms aggregation.

Example:

GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          { "product": { "terms": { "field": "product" } } }
        ]
      }
    }
  }
}

Like the terms aggregation, it’s possible to use a runtime field to create values for the composite buckets:

GET /_search
{
  "runtime_mappings": {
    "day_of_week": {
      "type": "keyword",
      "script": """
        emit(doc['timestamp'].value.dayOfWeekEnum
          .getDisplayName(TextStyle.FULL, Locale.ROOT))
      """
    }
  },
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          {
            "dow": {
              "terms": { "field": "day_of_week" }
            }
          }
        ]
      }
    }
  }
}

Histogram

edit

The histogram value source can be applied on numeric values to build fixed size interval over the values. The interval parameter defines how the numeric values should be transformed. For instance an interval set to 5 will translate any numeric values to its closest interval, a value of 101 would be translated to 100 which is the key for the interval between 100 and 105.

Example:

GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          { "histo": { "histogram": { "field": "price", "interval": 5 } } }
        ]
      }
    }
  }
}

Like the histogram aggregation it’s possible to use a runtime field to create values for the composite buckets:

GET /_search
{
  "runtime_mappings": {
    "price.discounted": {
      "type": "double",
      "script": """
        double price = doc['price'].value;
        if (doc['product'].value == 'mad max') {
          price *= 0.8;
        }
        emit(price);
      """
    }
  },
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          {
            "price": {
              "histogram": {
                "interval": 5,
                "field": "price.discounted"
              }
            }
          }
        ]
      }
    }
  }
}

Date histogram

edit

The date_histogram is similar to the histogram value source except that the interval is specified by date/time expression:

GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          { "date": { "date_histogram": { "field": "timestamp", "calendar_interval": "1d" } } }
        ]
      }
    }
  }
}

The example above creates an interval per day and translates all timestamp values to the start of its closest intervals. Available expressions for interval: year, quarter, month, week, day, hour, minute, second

Time values can also be specified via abbreviations supported by time units parsing. Note that fractional time values are not supported, but you can address this by shifting to another time unit (e.g., 1.5h could instead be specified as 90m).

Format

Internally, a date is represented as a 64 bit number representing a timestamp in milliseconds-since-the-epoch. These timestamps are returned as the bucket keys. It is possible to return a formatted date string instead using the format specified with the format parameter:

GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          {
            "date": {
              "date_histogram": {
                "field": "timestamp",
                "calendar_interval": "1d",
                "format": "yyyy-MM-dd"         
              }
            }
          }
        ]
      }
    }
  }
}

Supports expressive date format pattern

Time Zone

Date-times are stored in Elasticsearch in UTC. By default, all bucketing and rounding is also done in UTC. The time_zone parameter can be used to indicate that bucketing should use a different time zone.

Time zones may either be specified as an ISO 8601 UTC offset (e.g. +01:00 or -08:00) or as a timezone id, an identifier used in the TZ database like America/Los_Angeles.

Offset

Use the offset parameter to change the start value of each bucket by the specified positive (+) or negative offset (-) duration, such as 1h for an hour, or 1d for a day. See Time units for more possible time duration options.

For example, when using an interval of day, each bucket runs from midnight to midnight. Setting the offset parameter to +6h changes each bucket to run from 6am to 6am:

PUT my-index-000001/_doc/1?refresh
{
  "date": "2015-10-01T05:30:00Z"
}

PUT my-index-000001/_doc/2?refresh
{
  "date": "2015-10-01T06:30:00Z"
}

GET my-index-000001/_search?size=0
{
  "aggs": {
    "my_buckets": {
      "composite" : {
        "sources" : [
          {
            "date": {
              "date_histogram" : {
                "field": "date",
                "calendar_interval": "day",
                "offset": "+6h",
                "format": "iso8601"
              }
            }
          }
        ]
      }
    }
  }
}

Instead of a single bucket starting at midnight, the above request groups the documents into buckets starting at 6am:

{
  ...
  "aggregations": {
    "my_buckets": {
      "after_key": { "date": "2015-10-01T06:00:00.000Z" },
      "buckets": [
        {
          "key": { "date": "2015-09-30T06:00:00.000Z" },
          "doc_count": 1
        },
        {
          "key": { "date": "2015-10-01T06:00:00.000Z" },
          "doc_count": 1
        }
      ]
    }
  }
}

The start offset of each bucket is calculated after time_zone adjustments have been made.

GeoTile grid

edit

The geotile_grid value source works on geo_point fields and groups points into buckets that represent cells in a grid. The resulting grid can be sparse and only contains cells that have matching data. Each cell corresponds to a map tile as used by many online map sites. Each cell is labeled using a "{zoom}/{x}/{y}" format, where zoom is equal to the user-specified precision.

GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          { "tile": { "geotile_grid": { "field": "location", "precision": 8 } } }
        ]
      }
    }
  }
}

Precision

The highest-precision geotile of length 29 produces cells that cover less than 10cm by 10cm of land. This precision is uniquely suited for composite aggregations as each tile does not have to be generated and loaded in memory.

See Zoom level documentation on how precision (zoom) correlates to size on the ground. Precision for this aggregation can be between 0 and 29, inclusive.

Bounding box filtering

The geotile source can optionally be constrained to a specific geo bounding box, which reduces the range of tiles used. These bounds are useful when only a specific part of a geographical area needs high precision tiling.

GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          {
            "tile": {
              "geotile_grid": {
                "field": "location",
                "precision": 22,
                "bounds": {
                  "top_left": "52.4, 4.9",
                  "bottom_right": "52.3, 5.0"
                }
              }
            }
          }
        ]
      }
    }
  }
}

Mixing different value sources

edit

The sources parameter accepts an array of value sources. It is possible to mix different value sources to create composite buckets. For example:

GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          { "date": { "date_histogram": { "field": "timestamp", "calendar_interval": "1d" } } },
          { "product": { "terms": { "field": "product" } } }
        ]
      }
    }
  }
}

This will create composite buckets from the values created by two value sources, a date_histogram and a terms. Each bucket is composed of two values, one for each value source defined in the aggregation. Any type of combinations is allowed and the order in the array is preserved in the composite buckets.

GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          { "shop": { "terms": { "field": "shop" } } },
          { "product": { "terms": { "field": "product" } } },
          { "date": { "date_histogram": { "field": "timestamp", "calendar_interval": "1d" } } }
        ]
      }
    }
  }
}

Order

edit

By default the composite buckets are sorted by their natural ordering. Values are sorted in ascending order of their values. When multiple value sources are requested, the ordering is done per value source, the first value of the composite bucket is compared to the first value of the other composite bucket and if they are equals the next values in the composite bucket are used for tie-breaking. This means that the composite bucket [foo, 100] is considered smaller than [foobar, 0] because foo is considered smaller than foobar. It is possible to define the direction of the sort for each value source by setting order to asc (default value) or desc (descending order) directly in the value source definition. For example:

GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          { "date": { "date_histogram": { "field": "timestamp", "calendar_interval": "1d", "order": "desc" } } },
          { "product": { "terms": { "field": "product", "order": "asc" } } }
        ]
      }
    }
  }
}

... will sort the composite bucket in descending order when comparing values from the date_histogram source and in ascending order when comparing values from the terms source.

Missing bucket

edit

By default documents without a value for a given source are ignored. It is possible to include them in the response by setting missing_bucket to true (defaults to false):

GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          { "product_name": { "terms": { "field": "product", "missing_bucket": true } } }
        ]
      }
    }
  }
}

In the example above the source product_name will emit an explicit null value for documents without a value for the field product. The order specified in the source dictates whether the null values should rank first (ascending order, asc) or last (descending order, desc).

Size

edit

The size parameter can be set to define how many composite buckets should be returned. Each composite bucket is considered as a single bucket, so setting a size of 10 will return the first 10 composite buckets created from the value sources. The response contains the values for each composite bucket in an array containing the values extracted from each value source. Defaults to 10.

Pagination

edit

If the number of composite buckets is too high (or unknown) to be returned in a single response it is possible to split the retrieval in multiple requests. Since the composite buckets are flat by nature, the requested size is exactly the number of composite buckets that will be returned in the response (assuming that they are at least size composite buckets to return). If all composite buckets should be retrieved it is preferable to use a small size (100 or 1000 for instance) and then use the after parameter to retrieve the next results. For example:

GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "size": 2,
        "sources": [
          { "date": { "date_histogram": { "field": "timestamp", "calendar_interval": "1d" } } },
          { "product": { "terms": { "field": "product" } } }
        ]
      }
    }
  }
}

... returns:

{
  ...
  "aggregations": {
    "my_buckets": {
      "after_key": {
        "date": 1494288000000,
        "product": "mad max"
      },
      "buckets": [
        {
          "key": {
            "date": 1494201600000,
            "product": "rocky"
          },
          "doc_count": 1
        },
        {
          "key": {
            "date": 1494288000000,
            "product": "mad max"
          },
          "doc_count": 2
        }
      ]
    }
  }
}

To get the next set of buckets, resend the same aggregation with the after parameter set to the after_key value returned in the response. For example, this request uses the after_key value provided in the previous response:

GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "size": 2,
        "sources": [
          { "date": { "date_histogram": { "field": "timestamp", "calendar_interval": "1d", "order": "desc" } } },
          { "product": { "terms": { "field": "product", "order": "asc" } } }
        ],
        "after": { "date": 1494288000000, "product": "mad max" } 
      }
    }
  }
}

Should restrict the aggregation to buckets that sort after the provided values.

The after_key is usually the key to the last bucket returned in the response, but that isn’t guaranteed. Always use the returned after_key instead of derriving it from the buckets.

Early termination

edit

For optimal performance the index sort should be set on the index so that it matches parts or fully the source order in the composite aggregation. For instance the following index sort:

PUT my-index-000001
{
  "settings": {
    "index": {
      "sort.field": [ "username", "timestamp" ],   
      "sort.order": [ "asc", "desc" ]              
    }
  },
  "mappings": {
    "properties": {
      "username": {
        "type": "keyword",
        "doc_values": true
      },
      "timestamp": {
        "type": "date"
      }
    }
  }
}

This index is sorted by username first then by timestamp.

…​ in ascending order for the username field and in descending order for the timestamp field.

  1. could be used to optimize these composite aggregations:
GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          { "user_name": { "terms": { "field": "user_name" } } }     
        ]
      }
    }
  }
}

user_name is a prefix of the index sort and the order matches (asc).

GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          { "user_name": { "terms": { "field": "user_name" } } }, 
          { "date": { "date_histogram": { "field": "timestamp", "calendar_interval": "1d", "order": "desc" } } } 
        ]
      }
    }
  }
}

user_name is a prefix of the index sort and the order matches (asc).

timestamp matches also the prefix and the order matches (desc).

In order to optimize the early termination it is advised to set track_total_hits in the request to false. The number of total hits that match the request can be retrieved on the first request and it would be costly to compute this number on every page:

GET /_search
{
  "size": 0,
  "track_total_hits": false,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          { "user_name": { "terms": { "field": "user_name" } } },
          { "date": { "date_histogram": { "field": "timestamp", "calendar_interval": "1d", "order": "desc" } } }
        ]
      }
    }
  }
}

Note that the order of the source is important, in the example below switching the user_name with the timestamp would deactivate the sort optimization since this configuration wouldn’t match the index sort specification. If the order of sources do not matter for your use case you can follow these simple guidelines:

  • Put the fields with the highest cardinality first.
  • Make sure that the order of the field matches the order of the index sort.
  • Put multi-valued fields last since they cannot be used for early termination.

index sort can slowdown indexing, it is very important to test index sorting with your specific use case and dataset to ensure that it matches your requirement. If it doesn’t note that composite aggregations will also try to early terminate on non-sorted indices if the query matches all document (match_all query).

Sub-aggregations

edit

Like any multi-bucket aggregations the composite aggregation can hold sub-aggregations. These sub-aggregations can be used to compute other buckets or statistics on each composite bucket created by this parent aggregation. For instance the following example computes the average value of a field per composite bucket:

GET /_search
{
  "size": 0,
  "aggs": {
    "my_buckets": {
      "composite": {
        "sources": [
          { "date": { "date_histogram": { "field": "timestamp", "calendar_interval": "1d", "order": "desc" } } },
          { "product": { "terms": { "field": "product" } } }
        ]
      },
      "aggregations": {
        "the_avg": {
          "avg": { "field": "price" }
        }
      }
    }
  }
}

... returns:

{
  ...
  "aggregations": {
    "my_buckets": {
      "after_key": {
        "date": 1494201600000,
        "product": "rocky"
      },
      "buckets": [
        {
          "key": {
            "date": 1494460800000,
            "product": "apocalypse now"
          },
          "doc_count": 1,
          "the_avg": {
            "value": 10.0
          }
        },
        {
          "key": {
            "date": 1494374400000,
            "product": "mad max"
          },
          "doc_count": 1,
          "the_avg": {
            "value": 27.0
          }
        },
        {
          "key": {
            "date": 1494288000000,
            "product": "mad max"
          },
          "doc_count": 2,
          "the_avg": {
            "value": 22.5
          }
        },
        {
          "key": {
            "date": 1494201600000,
            "product": "rocky"
          },
          "doc_count": 1,
          "the_avg": {
            "value": 10.0
          }
        }
      ]
    }
  }
}

Pipeline aggregations

edit

The composite agg is not currently compatible with pipeline aggregations, nor does it make sense in most cases. E.g. due to the paging nature of composite aggs, a single logical partition (one day for example) might be spread over multiple pages. Since pipeline aggregations are purely post-processing on the final list of buckets, running something like a derivative on a composite page could lead to inaccurate results as it is only taking into account a "partial" result on that page.

Pipeline aggs that are self contained to a single bucket (such as bucket_selector) might be supported in the future.